CN108242182A - Information processing unit, information processing method and program - Google Patents
Information processing unit, information processing method and program Download PDFInfo
- Publication number
- CN108242182A CN108242182A CN201711390135.3A CN201711390135A CN108242182A CN 108242182 A CN108242182 A CN 108242182A CN 201711390135 A CN201711390135 A CN 201711390135A CN 108242182 A CN108242182 A CN 108242182A
- Authority
- CN
- China
- Prior art keywords
- curved mirror
- information
- vehicle
- image
- eyes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 107
- 238000003672 processing method Methods 0.000 title claims abstract description 24
- 238000004364 calculation method Methods 0.000 claims abstract description 132
- 238000001514 detection method Methods 0.000 claims abstract description 49
- 230000008859 change Effects 0.000 claims description 52
- 230000009471 action Effects 0.000 claims description 45
- 238000012360 testing method Methods 0.000 claims description 35
- 230000002123 temporal effect Effects 0.000 claims description 22
- 238000000034 method Methods 0.000 abstract description 43
- 230000008569 process Effects 0.000 abstract description 21
- 238000012545 processing Methods 0.000 description 184
- 239000000203 mixture Substances 0.000 description 21
- 230000006870 function Effects 0.000 description 18
- 238000004590 computer program Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 12
- 238000013135 deep learning Methods 0.000 description 10
- 206010039203 Road traffic accident Diseases 0.000 description 7
- 230000000694 effects Effects 0.000 description 7
- 238000010276 construction Methods 0.000 description 4
- 230000003595 spectral effect Effects 0.000 description 4
- 230000004308 accommodation Effects 0.000 description 3
- 230000002547 anomalous effect Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 238000007689 inspection Methods 0.000 description 2
- 229910000530 Gallium indium arsenide Inorganic materials 0.000 description 1
- 206010027146 Melanoderma Diseases 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 230000001351 cycling effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000004438 eyesight Effects 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000000178 monomer Substances 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- VMXUWOKSQNHOCA-UKTHLTGXSA-N ranitidine Chemical compound [O-][N+](=O)\C=C(/NC)NCCSCC1=CC=C(CN(C)C)O1 VMXUWOKSQNHOCA-UKTHLTGXSA-N 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
- B60R2300/60—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by monitoring and displaying vehicle exterior scenes from a transformed perspective
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
- B60R2300/80—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the intended use of the viewing arrangement
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
- B60R2300/80—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the intended use of the viewing arrangement
- B60R2300/8033—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the intended use of the viewing arrangement for pedestrian protection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
- G06T2207/30256—Lane; Road marking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
- G06T2207/30261—Obstacle
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/167—Driving aids for lane monitoring, lane changing, e.g. blind spot detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Aviation & Aerospace Engineering (AREA)
- Electromagnetism (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Psychiatry (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
Being designed to provide for the disclosure a kind of can look into the distance information processing unit, information processing method and program that the safe driving of vehicle is supported in the place of condition difference in intersection etc..Information processing unit has：The curved mirror appeared before one's eyes in the picture and the object appeared before one's eyes in curved mirror based on the image information of image that expression is obtained by the shooting for the filming apparatus for being equipped on vehicle, detect in detection process portion (12a)；Calculation section (101A) calculates the position of the object in detected curved mirror；Generating unit generates drive supporting information based on the position of the object in the curved mirror calculated；And output section, the generated drive supporting information of output.
Description
Technical field
This disclosure relates to information processing unit, information processing method and program.
Background technology
The place of condition (watching from a height or a distance condition) difference is looked into the distance for the driver of vehicle in intersection etc., is easily occurred for example
The such accident contacted with other vehicles or pedestrian, collided.Therefore, waiting in expectation exploitation can be in intersection etc.
The technology that the danger that place can such as be related to occurring accident is predicted, can prevent accident from occurring.
For example, patent document 1 discloses following technology：Make vehicle more by applying flexibly the data recorded by drive recorder
Safely travel.In the patent document 1, record represents the row when anomalous event has occurred in vehicle in the drive recorder
Sail the data of situation.Also, the driving condition that the data recorded in predicting by drive recorder are grasped can be reproduced
When, arouse driver and pay attention to.In such manner, it is possible to safety is improved, therefore, it is possible to which vehicle is made more safely to travel.
Existing technical literature
Patent document 1：Japanese Unexamined Patent Publication 2007-193577 bulletins
Invention content
Technical problems to be solved by the inivention
But predicted in dangerous method directly applying flexibly the data that drive recorder observed, due to being related to
The time interval that the anomalous event of hidden danger occurs is short, so being difficult to fully record the row represented when the anomalous event occurs sometimes
Sail the data of situation.Accordingly, it is difficult to predict the driving condition reproduction in addition can it is difficult to predict such as relating to the danger that accident occurs
Danger.In other words, it is above-mentioned in the prior art, it is difficult to prediction can be related to occurring dangerous as accident, it is difficult to carry out control vehicle,
Arouse the drive supporting that driver pays attention to etc..Especially, in the place for looking into the distance condition difference, the observation condition of the narrow grade of observation scope is disliked
Change, therefore, deficiency easily occurs for the data of drive recorder observation.
The disclosure is completed in view of the above circumstances, and it is an object of the present invention to provide a kind of can look into the distance item in intersection etc.
Information processing unit that the safe driving of vehicle is supported in the place of part difference etc..
For solving the technical solution of technical problem
To achieve these goals, the information processing unit that a technical solution of the disclosure is related to has：Test section, base
In the image information of image that expression is obtained by the shooting for the filming apparatus for being equipped on vehicle, detect and appear before one's eyes in described image
Curved mirror and the object appeared before one's eyes in the curved mirror；Calculation section calculates the institute in the detected curved mirror
State the position of object；Generating unit, based on the position of the object in the curved mirror calculated, generation drive supporting letter
Breath；And output section, export the drive supporting information generated.
In addition, these master or specific technical solutions can pass through system, method, integrated circuit, computer
The recording mediums such as program or computer-readable CD-ROM are realized, can also pass through system, method, integrated circuit, meter
Calculation machine program and the arbitrary of recording medium combine to realize.
The effect of invention
According to the disclosure, it can realize that the place that condition difference can be looked into the distance in intersection etc. carries out the safe driving of vehicle
Information processing unit of support etc..
Description of the drawings
Fig. 1 is the block diagram of an example of the composition for the system for representing embodiment 1.
Fig. 2 is the figure of an example that the function for the image acquiring section for representing embodiment 1 is formed.
Fig. 3 is the figure for an example for representing the situation when image acquiring section of embodiment 1 is equipped on vehicle.
Fig. 4 is the definition graph of an example of the image acquired by the image acquiring section of embodiment 1.
Fig. 5 is the definition graph of the curved mirror detection of the detection based on simple circle or rectangle etc..
Fig. 6 is the definition graph of the curved mirror detection of the image identification based on statistics.
Fig. 7 be utilized deep learning (Deep Learning) in the identification side of moving object that curved mirror is appeared before one's eyes
The definition graph of method.
Fig. 8 is the figure of an example that the function for the calculation section for representing embodiment 1 is formed.
The definition graph of an example of calculating processing that the calculation section of the expression embodiment 1 of Fig. 9 carries out.
Figure 10 is to represent the definition graph of an example of dead angle determination processing that the dead angle determination unit of embodiment 1 carries out.
Figure 11 is the definition graph at the dead angle of curved mirror.
Figure 12 is the figure of an example that the function for the output processing part for representing embodiment 1 is formed.
Figure 13 is the schematic diagram of the variable quantity for the size for representing the object in Fig. 9 (b).
Figure 14 is the figure of an example for the curve turning road for representing embodiment 1.
Figure 15 is to represent the definition graph of an example of output processing that the output processing part of embodiment 1 carries out.
Figure 16 is the figure of an example for the T words intersection for representing embodiment 1.
Figure 17 is to represent the definition graph of other an examples of output processing that the output processing part of embodiment 1 carries out.
Figure 18 is the flow chart of the summary of the information processing method for the system for representing embodiment 1.
Figure 19 is the flow chart of the details of the information processing method for the system for representing embodiment 1.
Figure 20 is saying for an example of risk prediction processing that the output processing part for the variation 1 for representing embodiment 1 carries out
Bright figure.
Figure 21 is saying for an example of risk prediction processing that the output processing part for the variation 1 for representing embodiment 1 carries out
Bright figure.
Figure 22 is saying for an example of risk prediction processing that the output processing part for the variation 2 for representing embodiment 1 carries out
Bright figure.
Figure 23 is saying for an example of risk prediction processing that the output processing part for the variation 2 for representing embodiment 1 carries out
Bright figure.
Figure 24 is saying for an example of risk prediction processing that the output processing part for the variation 3 for representing embodiment 1 carries out
Bright figure.
Figure 25 is saying for an example of risk prediction processing that the output processing part for the variation 3 for representing embodiment 1 carries out
Bright figure.
Figure 26 is the block diagram of an example of the composition for the system for representing embodiment 2.
Figure 27 A be the vehicle for representing to temporarily cease in intersection driver's visual confirmation to curved mirror in appeared before one's eyes
Object an example figure.
Figure 27 B are the definition graphs of the characteristic of curved mirror.
Figure 27 C are the definition graphs of the characteristic of curved mirror.
Figure 28 is the figure of an example that the function for the calculation section for representing embodiment 2 is formed.
Figure 29 A are to represent the figure of an example of the position of object that the object space calculation section shown in Figure 28 calculates.
Figure 29 B are to represent the figure of an example in road surface region that the road surface region calculation section shown in Figure 28 calculates.
Figure 30 is the definition graph of the calculation method of central shaft that the central shaft calculation section shown in Figure 28 calculates.
Figure 31 is the figure of an example that the function for the output processing part for representing embodiment 2 is formed.
Figure 32 is the figure of an example of the position of the object in the curved mirror for represent embodiment 2.
Figure 33 is the definition graph of the danger level in the case that the object of embodiment 2 is located at the inboard of curved mirror.
Figure 34 is the definition graph of the danger level in the case that the object of embodiment 2 is located at the inboard of curved mirror.
Figure 35 is the definition graph of the danger level in the case that the object of embodiment 2 is located at the inboard of curved mirror.
Figure 36 is the definition graph of the danger level in the case that the object of embodiment 2 is located at the front side of curved mirror.
Figure 37 is the definition graph of the danger level in the case that the object of embodiment 2 is located at the front side of curved mirror.
Figure 38 is the definition graph of the danger level in the case that the object of embodiment 2 is located at the front side of curved mirror.
Figure 39 is the figure of an example for the T words intersection for representing embodiment 2.
Figure 40 is to represent the definition graph of other an examples of output processing that the output processing part of embodiment 2 carries out.
Figure 41 is the flow chart of the summary of the information processing method for the system for representing embodiment 2.
Figure 42 is the flow chart of the details of the information processing method for the system for representing embodiment 2.
Figure 43 is saying for an example of risk prediction processing that the output processing part for the variation 1 for representing embodiment 2 carries out
Bright figure.
Figure 44 is saying for an example of risk prediction processing that the output processing part for the variation 2 for representing embodiment 2 carries out
Bright figure.
Figure 45 is saying for an example of risk prediction processing that the output processing part for the variation 3 for representing embodiment 2 carries out
Bright figure.
Figure 46 is the block diagram of an example of the composition for the system for representing embodiment 3.
Figure 47 is the figure of an example that the function for the feature determination unit for representing embodiment 3 is formed.
Figure 48 is the definition graph for an example for representing Dynamic Graph.
Figure 49 A are an examples of the ambient enviroment for the curved mirror that the image of embodiment 3 is appeared before one's eyes.
Figure 49 B are an examples of the ambient enviroment for the curved mirror that the image of embodiment 3 is appeared before one's eyes.
Figure 49 C are an examples of the ambient enviroment for the curved mirror that the image of embodiment 3 is appeared before one's eyes.
Figure 49 D are an examples of the ambient enviroment for the curved mirror that the image of embodiment 3 is appeared before one's eyes.
Figure 49 E are an examples of the ambient enviroment for the curved mirror that the image of embodiment 3 is appeared before one's eyes.
Figure 50 is the figure for representing an example of map used in the setting place acquisition unit of embodiment 3.
Figure 51 is to represent figure of the 1st danger level determination unit of embodiment 3 for an example of the local Dynamic Graph of judgement.
Figure 52 is to represent figure of the 1st danger level determination unit of embodiment 3 for an example of the local Dynamic Graph of judgement.
Figure 53 is to represent figure of the 1st danger level determination unit of embodiment 3 for an example of the local Dynamic Graph of judgement.
Figure 54 is to represent figure of the 1st danger level determination unit of embodiment 3 for an example of the local Dynamic Graph of judgement.
Figure 55 is the figure of an example that the function for the output processing part for representing embodiment 3 is formed.
Figure 56 is to represent the definition graph of an example of output processing that the output processing part of embodiment 3 carries out.
Figure 57 is to represent the definition graph of an example of output processing that the output processing part of embodiment 3 carries out.
Figure 58 is the flow chart of the summary of the information processing method for the system for representing embodiment 3.
Figure 59 is the flow chart of the details of the information processing method for the system for representing embodiment 3.
Figure 60 is the 2nd danger level determination processing that the 2nd danger level determination unit of the variation 1 for representing embodiment 3 carries out
An example definition graph.
Figure 61 is the 2nd danger level determination processing that the 2nd danger level determination unit of the variation 2 for representing embodiment 3 carries out
An example definition graph.
Figure 62 is the 2nd danger level determination processing that the 2nd danger level determination unit of the variation 3 for representing embodiment 3 carries out
An example definition graph.
Label declaration
1st, 1A, 1B system
2nd, 2A, 3 vehicles
10th, 10A, 10B information processing unit
11 image acquiring sections
11B acquisition units
12nd, 12B test sections
12a, 12b detection process portion
13rd, 13B identification parts
50、50a、50j、50j+1、50j+2、50j+3、50k、50k+1、50k+2、50k+3、50m、50m+1、50m+2、50m+3、50n、
50n+1、50n+2Image
51、51m、51m+1、51m+2、51n+1、51n+2、51′、51′n+1、51′n+2, 71,72,73,74,75 curved mirrors
51A edges
52 mirror shape filters
52AMatching area
53rd, 53a positions
54th, 54a similarity graphs
60、60m、60m+1、60m+2、60n+1、60n+2、60′、60′n+1、60′n+2People
61m+1、61m+2Moving object
64、64m、64m+1、64m+2Object
80 road surfaces region
82 central shafts
86 maps
86t、86t1、86t2、86t3Local Dynamic Graph
101st, 101A calculation sections
102nd, 102A, 102B output processing part
103 dead angle determination units
104 determination units
111 detecting parts
112 image recording portions
860th, 865 commercial facility
861st, 862,864 road
863 super expressways
866 concerts
1011 difference calculation sections
1012nd, 1014 dimension of object calculation section
1013 object space calculation sections
1015 road surface region calculation sections
1016 central shaft calculation sections
1031st, 1031A risk predictions portion
1032nd, 1032A, 1032B information generation unit
1033rd, 1033A, 1033B information output part
1041 feature determination units
1042 the 1st danger level determination units
1043 the 2nd danger level determination units
Specific embodiment
The information processing unit that one technical solution of the disclosure is related to has：Test section is equipped on based on representing to pass through
The image information of image that the shooting of the filming apparatus of vehicle obtains detects the curved mirror appeared before one's eyes in described image and in institute
State the object appeared before one's eyes in curved mirror；Calculation section calculates the position of the object in the detected curved mirror；Generation
Portion based on the position of the object in the curved mirror calculated, generates drive supporting information；And output section,
Export the drive supporting information generated.
Thereby, it is possible to the driving branch of vehicle is exported by looking into the distance the place of condition difference in intersection etc. using curved mirror
Information is helped, therefore, it is possible to support the safe driving of vehicle.
Here, for example or, the generating unit is true according to the object and the road by appearing before one's eyes in described image
The position relationship between region in the fixed curved mirror, generates the drive supporting information.
In addition, for example or, the calculation section has：Object space calculation section is calculated in the curved mirror
The position of the object；Road surface region calculation section calculates the road surface region in the curved mirror；And central shaft calculation section,
Its central shaft in the road surface region calculated, the generating unit is according to the institute calculated by the object space calculation section
The position of object is stated between the central shaft that is calculated by the central shaft calculation section position relationship generates the driving
Support information.
In addition, for example or, the generating unit is located in the object in described image to be leaned on than the central shaft
On position in the case of, generate the drive supporting information for the vehicle to be made to hide the object.
Thereby, it is possible to hide object in the case where object is near away from vehicle, it can be ensured that the safety of object and vehicle.
In addition, for example or, the generating unit is based on the position calculated by the calculation section and the curved surface
Position relationship between the center line of mirror generates the drive supporting information.
In addition, for example or, the generating unit, (i) is in the feelings on curved mirror direction left side from the point of view of the vehicle
When the object is located at the position kept right than the center line in described image under condition, alternatively, (ii) the curved mirror from
It is raw when being located at the position to keep left than the center line in described image towards the object in the case of right from the point of view of the vehicle
Into for the vehicle to be made to hide the drive supporting information of the object.
Thereby, it is possible to hide object in the case where object is near away from vehicle, it can be ensured that the safety of object and vehicle.
In addition, the complicated processing of the central shaft as determined road can be saved, processing speed can be improved.
In addition, for example or, the drive supporting information is according to the vehicle to the direction towards the curved mirror
The danger of position prediction under the situation of traveling, by the object in the curved mirror exports.
In addition, for example or, the calculation section is also equipped with dimension of object calculation section, the dimension of object calculation section is calculated
Go out the size of the object on the basis of the curved mirror, the generating unit is based on the object calculated in the curved surface
Position in mirror and the size of the object that is calculated generate the drive supporting information.
Here, for example or, the generating unit is according to the respective middle institute of continuous at least two image in temporal sequence
Position of the change information and the object of the respective size of the object appeared before one's eyes in the curved mirror is driven described in generation
Sail support information.
In addition, for example or, the drive supporting information includes the control instruction information of the movement of the vehicle.
In addition, for example or, the drive supporting information includes the prompt message prompted to the occupant of the vehicle.
In addition, for example or, described information processing unit has the object that identification is appeared before one's eyes in the curved mirror
Identification part.
In addition, for example or, the generating unit according to the attribute of the object recognized by the identification part come
Generate the drive supporting information.
In addition, for example or, the generating unit is behaved in the attribute of the object recognized by the identification part
In the case of object, according to the information related with the age of the object recognized by the identification part, make the drive supporting
Information change and generate the drive supporting information.
In addition, for example or, the generating unit is behaved in the attribute of the object recognized by the identification part
In the case of object, whether do not paid attention to taking action according to the object recognized by the identification part, make the driving
It supports information change and generates the drive supporting information.
In addition, for example or, the generating unit further in continuous multiple images in temporal sequence at least
In one image by the object that the identification part recognizes than at least one image in time series rearward
In the case of not recognized in image by the identification part, the drive supporting information is generated.
In addition, for example or, the generating unit based on the position of the object appeared before one's eyes in one image,
This case is not recognized by the identification part with the object described in the image rearward, generates the drive supporting letter
Breath.
The information processing method that one technical solution of the disclosure is related to is performed using processor：It is equipped on based on representing to pass through
The image information of image that the shooting of the filming apparatus of vehicle obtains detects the curved mirror appeared before one's eyes in described image and in institute
State the object appeared before one's eyes in curved mirror；Calculate the position of the object in the detected curved mirror；Based on what is calculated
The position of the object in the curved mirror generates drive supporting information；Export the drive supporting information generated.
The program that one technical solution of the disclosure is related to is to carry out the program that the computer of information processing method can be read,
Described program causes：Based on the image information of image that expression is obtained by the shooting for the filming apparatus for being equipped on vehicle, detection
The curved mirror appeared before one's eyes in described image and the object appeared before one's eyes in the curved mirror；Calculate the detected curved mirror
In the object position；Based on the position of the object in the curved mirror calculated, drive supporting information is generated；
Export the drive supporting information generated.
Hereinafter, embodiment of the present disclosure is illustrated with reference to attached drawing.
Embodiments described below all represents a concrete example of the disclosure.The numerical value that is shown in following embodiment,
Shape, inscape, step, sequence of step etc. are an examples, and the meaning of the non-limiting disclosure.In addition, about following reality
The inscape not being documented in the independent claims for representing upper concept in the inscape in mode is applied, is conduct
Arbitrary inscape and illustrate.Alternatively, it is also possible to combine respective content in all embodiments.
(embodiment 1)
[composition of system 1]
Fig. 1 is the block diagram of an example of the composition for the system 1 for representing embodiment 1.
The vehicle such as being equipped on automobile of system 1 shown in FIG. 1 has information processing unit 10, image acquiring section 11, inspection
Processing unit 12a is surveyed, exports the drive supporting information of the vehicle.The details of these compositions etc. is illustrated below.In addition,
Image acquiring section 11 and detection process portion 12a are not limited to situation shown in FIG. 1, can also be located at information processing unit 10.
Here, can mostly the place for looking into the distance condition difference be provided with curved mirror.For example, curved mirror is true for the safety of auxiliary
The convex mirror recognized, be arranged on road so that look into the distance the intersection of condition difference or be difficult to see that front turning,
The situation that can visually cannot be directly viewed.Then, information processing unit 10 carrys out the more peace to the vehicle by using curved mirror
Full driving is supported.In addition, following curved mirror has physically existing round shape or rectangular-shaped mirror, but also may be used
The round shape for the situation that cannot be directly viewed or rectangular-shaped electron mirror are shown by image to have.
[image acquiring section 11]
Fig. 2 is the figure of an example that the function for the image acquiring section 11 for representing embodiment 1 is formed.Fig. 3 is to represent embodiment party
The figure of an example of situation when the image acquiring section 11 of formula 1 is equipped on vehicle.Fig. 4 is that the image acquiring section 11 of embodiment 1 takes
The definition graph of an example of the image obtained.
The image that image acquiring section 11 obtains the image for representing to obtain by the shooting for the filming apparatus for being equipped on vehicle is believed
Breath.In the present embodiment, as shown in Fig. 2, image acquiring section 11 has detecting part 111 and image recording portion 112.For example, figure
As acquisition unit 11 is vehicle-mounted camera as shown in Figure 3, it is equipped on vehicle.
Sensing (sensing) portion 111 shoots image in temporal sequence and continuously to the front of the vehicle during traveling, and
It is recorded in image recording portion 112.Here, detecting part 111 is the moving picture recording apparatus of camera etc..In addition, sensing
Portion 111 can also be the visible image capturing head for carrying out visible light shooting or the InGaAs cameras for carrying out infrared light shooting.
In this case, detecting part 111 shoots image on daytime with visible ray, image is shot with infrared light at night.
Image recording portion 112 has HDD (Hard Disk Drive, hard disk drive) or memory etc., record sensing
Image taken by portion 111.
Hereinafter, image 50 for example, as shown in figure 4, is clapped with the place detecting part 111 that condition difference is looked into the distance in intersection etc.
It is illustrated in case of taking the photograph obtained from the front of the vehicle during advancing curved mirror 51 of appearing before one's eyes in image.
[detection process portion 12a]
Images of the detection process portion 12a based on the image for representing to obtain by the shooting for the filming apparatus for being equipped on vehicle is believed
Breath, is detected the curved mirror appeared before one's eyes in image and the object appeared before one's eyes in curved mirror.In addition, detection process portion 12a
It can also identify the attribute for the object appeared before one's eyes in curved mirror.In the present embodiment, detection process portion 12a has 12 He of test section
Identification part 13.
12 ＞ of ＜ test sections
The image information of image that test section 12 is obtained based on expression by the shooting for the filming apparatus for being equipped on vehicle is right
It is detected in the curved mirror that image is appeared before one's eyes.More specifically, test section 12 is detected by the vehicle during traveling
The curved mirror that image obtained from front is continuously shot in temporal sequence is appeared before one's eyes.In addition, shooting direction can also be vehicle
Front other than direction.For example, in the case where vehicle is rearward advanced, shooting direction can also be the rear of vehicle.
In the present embodiment, test section 12 is in the image 50 for example shown in Fig. 4 obtained by image acquiring section 11
In the curved mirror 51 appeared before one's eyes be detected.Here, the method for curved mirror in detection image for example including be based on simply circle or
The method of the detection of person's rectangle etc. or image based on statistics know method for distinguishing etc..Hereinafter, using figure come in detection image
The method of curved mirror illustrate.
Fig. 5 is the definition graph of the curved mirror detection of the detection based on simple circle or rectangle etc..In addition, pair same with Fig. 4
The element of sample marks identical label, and omits detailed description.
First, test section 12 carries out edge inspection in the image 50 for example shown in fig. 5 obtained by image acquiring section 11
It surveys.As a result, as shown in image 50a, test section 12 detects edge 51a corresponding with the shape of curved mirror.
Then, test section 12 carries out the search of edge similar degree in the image 50a for having carried out edge detection.It is more specific and
Speech, as shown in image 50a, test section 12 makes filter (the mirror shape filter of detection circle or rectangle etc. in image 50a
52) it is scanned, the search edge similar to mirror shape filter 52.In addition, mirror shape filter 52 is for detecting curved surface
The filter of the profile of mirror, such as Sobel filter (sobel filter), Canny filters (Canny can be used
Filter) etc..
Also, test section 12 for example makes similarity graph (map) 54, the high position of the similarity in detection similarity Figure 54
53 object is used as curved mirror 51.
In addition, the position on the map of subject that the image obtained by image acquiring section 11 is appeared before one's eyes is known
In the case of, can also the shape based on the curved mirror according to the position deduction etc. information, selection uses detection circle or square
The filter of shape etc..
Fig. 6 is the definition graph of the detection of the curved mirror of the image identification based on statistics.In addition, pair with the same elements of Fig. 4
Identical label is marked, and omits detailed description.
For example, test section 12 makes tool in the image 50 for example shown in (a) of Fig. 6 obtained by image acquiring section 11
Standby matching (matching) region 52a of circle for changing size is scanned.
As a result, as shown in (b) of Fig. 6, test section 12 makes the size by each position in image 50 and its position
Represent the similarity graph 54a with the similarity of the circle.
Also, test section 12 detects the object of the high position 53a of similarity in similarity graph 54a as curved mirror 51.
In addition, the image identification of statistics is not limited to situation about illustrating using Fig. 6.Test section 12 may be to have advance
The identifier (Deep Learning, SVM etc.) of the image other than the image and curved mirror of a large amount of curved mirror is practised.In the feelings
Under condition, test section 12 implements the image that each position in image 50 determines whether curved mirror using the identifier, thus
Detect curved mirror 51.
13 ＞ of ＜ identification parts
The image information of image that identification part 13 is obtained based on expression by the shooting for the filming apparatus for being equipped on vehicle is known
The object do not appeared before one's eyes in curved mirror.More specifically, identification part 13 is identified reflects in the image obtained by image acquiring section 11
Object in existing curved mirror.In addition, identification part 13 can also identify the attribute for the object appeared before one's eyes in curved mirror.
In the present embodiment, identification part 13 is identified in the curved mirror included in the image detected by test section 12
The positions and dimensions of the mobile objects such as people, bicycle or the automobile appeared before one's eyes (hereinafter also referred to moving object).Here, identification exists
The method of existing moving object is such as including the use of deep learning (Deep Learning) in the curved mirror that image is appeared before one's eyes
The method of machine learning.Hereinafter, using Fig. 7 to the method for existing moving object in the curved mirror appeared before one's eyes in image of identification into
Row explanation.
Fig. 7 be utilized deep learning (Deep Learning) in the identification side of moving object that curved mirror is appeared before one's eyes
The definition graph of method.In (a) and (b) of Fig. 7, show and appear before one's eyes that there are people 60 i.e. in curved mirror 51 by people 60 in curved mirror 51
In the case of example.In addition, pair label identical with the same element labels of Fig. 4, omits detailed description.
Identification part 13 carries out following identifying processing：For in the image 50 for example as shown in (a) of Fig. 7 by test section
The region of the curved mirror 51 of 12 detections, such as change position and size as shown in (b) of Fig. 7 on one side, on one side to others
60 are known.
In addition, in order to which identification part 13 is made to carry out above-mentioned identifying processing, moving object is identified to identification part 13 in advance
Study i.e. other than people, bicycle and automobile and moving object.If it is identified as the object being present in curved mirror as a result,
When the reliability of body is more than threshold value and the object there are size maximum, identification part 13 can be identified as appearing before one's eyes have fortune in curved mirror
Animal body.In addition, identification part 13 is being identified as, when curved mirror appears before one's eyes and has moving object, exporting its positions and dimensions.It here, can
It is represented by degree for representing people, the bicycle value that either shape of the moving object of automobile or profile characteristic are indicated.
On the other hand, if the reliability for being not identified as being present in the object in curved mirror is more than the object of threshold value, identification part 13 is known
Wei not appear before one's eyes in curved mirror has moving object.
[information processing unit 10]
Then, the information processing unit 10 of embodiment 1 is illustrated.
The information processing unit 10 of embodiment 1 exports the drive supporting information of the vehicle by using curved mirror.At this
In embodiment, as shown in Figure 1, information processing unit 10 has calculation section 101, output processing part 102, dead angle determination unit 103.
Hereinafter, the details of these compositions etc. is illustrated.
101 ＞ of ＜ calculation sections
Calculation section 101 is calculated through the detection process portion 12a sizes of curved mirror detected and the difference of reference dimension, is made
The size of object is calculated with the difference calculated.Here, difference includes multiplying power of the curved mirror relative to reference dimension.Object
Size corresponding with difference zoom in or out to calculate by carrying out the object appeared before one's eyes in curved mirror.
Fig. 8 is the figure of an example that the function for the calculation section 101 for representing embodiment 1 is formed.
As shown in figure 8, the calculation section 101 of embodiment 1 has difference calculation section 1011 and dimension of object calculation section 1012.
Difference calculation section 1011 calculates the size of the curved mirror detected by test section 12 and the difference of reference dimension.More
For body, difference calculation section 1011 calculate by test section 12 detect by the front to the vehicle during traveling temporally
Difference between the size and reference dimension of the curved mirror appeared before one's eyes in image obtained from shooting to Sequentially continuous.In this embodiment party
In formula, difference calculation section 1011 for example calculates the curved mirror appeared before one's eyes in the image as obtained from shooting relative to reference dimension
Multiplying power be used as difference.Reference dimension is the size of the height and width means by predetermined curved mirror.
In addition, reference dimension both can be in the curved mirror appeared before one's eyes in the multiple images that are taken in the predetermined time most
The size of big curved mirror, or the size of curved mirror appeared before one's eyes in the image before 1 taken frame.In addition, difference
Calculation section 1011 can also be calculated by being put according to the multiplying power calculated in continuous at least two image in temporal sequence
Big or the object obtained from reducing object in curved mirror size changing amount is used as above-mentioned difference.
In addition, difference calculation section 1011 can also calculate above-mentioned difference according to 1 image.Specifically, reference dimension is
Pre-stored value, size of the difference calculation section 1011 based on the curved mirror appeared before one's eyes in 1 image and the station meter stored
It is very little to calculate above-mentioned multiplying power.In addition, reference dimension can also be associated with the distance between subject, according to vehicle with
The distance between curved mirror appeared before one's eyes in the picture is corrected.
Dimension of object calculation section 1012 calculates the size of object using the difference calculated by difference calculation section 1011.More
Specifically, dimension of object calculation section 1012 is using the object appeared before one's eyes in the curved mirror recognized by identification part 13 and by difference
The difference that calculation section 1011 calculates calculates the size of the object.In the present embodiment, dimension of object calculation section 1012 is by making
The object appeared before one's eyes in the curved mirror recognized by identification part 13 according to the difference calculated by difference calculation section 1011 amplifying or
Person reduces the size to calculate object, the size as the object appeared before one's eyes in curved mirror.
Here, it is illustrated using attached drawing come an example handled the calculating that the calculation section 101 formed in this way carries out.
Fig. 9 is to represent the definition graph of an example of calculating processing that the calculation section 101 of embodiment 1 carries out.In addition, pair with figure
4 same elements mark identical label, omit detailed description.
Image 50 shown in (a) of Fig. 9n, image 50n+1And image 50n+2Be to the vehicle front during traveling by
T at the time of time series is continuousn, moment tn+1And moment tn+2The image taken.Image 50n, image 50n+1And image
50n+2Respectively comprising curved mirror 51, curved mirror 51n+1And curved mirror 51n+2, in curved mirror 51, curved mirror 51n+1And curved mirror
51n+2In appear before one's eyes someone 60, people 60n+1And people 60n+2.In addition, curved mirror 51, curved mirror 51n+1And curved mirror 51n+2Ruler
Very little is, for example, (w1a、h1a)、(w1b、h1b) and (w1c、h1c), it is showed by width and height.Curved mirror 51, curved mirror 51n+1
And curved mirror 51n+2Size detected by test section 12.Show to be amplified to the curved mirror of reference dimension at (b) of Fig. 9
51 ', curved mirror 51 'n+1And curved mirror 51 'n+2。
In this case, calculation section 101 is that the calculating of difference calculation section 1011 will be in image 50n, image 50n+1And image
50n+2In curved mirror 51, the curved mirror 51 appeared before one's eyesn+1And curved mirror 51n+2Zoom in or out multiplying power during reference dimension.
Reference dimension is for example set as (ws、hs) in the case of, difference calculation section 1011 is calculated curved mirror 51,51n+1And 51n+2It puts
Multiplying power when greatly or narrowing down to reference dimension is ws/w1a、ws/w1b、ws/w1c.It can also calculate as hs/h1a、hs/h1b、hs/
h1c。
Also, as shown in (b) of Fig. 9, calculation section 101 is that dimension of object calculation section 1012 is calculated according to by difference calculation section
The 1011 multiplying power w calculateds/w1a、ws/w1b、ws/w1cMake curved mirror 51, curved mirror 51n+1And curved mirror 51n+2, people 60, people
60n+1And people 60n+2People 60 ', people 60 ' during amplificationn+1And people 60 'n+2Size.
103 ＞ of ＜ dead angles determination unit
Dead angle determination unit 103 is at least one image of continuous multiple images in temporal sequence by detection process portion
12a detect (identification) to object in than image of at least one image in time series rearward not by detection process portion
It is the dead of the dead angle that the object is likely to be present in curved mirror by the spectral discrimination rearward in the case that 12a detections (identification) are arrived
Angle image.In other words, dead angle determination unit 103 judges what whether can be seen in curved mirror in continuous image in time series
Moving object can become can't see, and will become the spectral discrimination that can't see as dead angle image.
Here, it is said using an example of dead angle determination processing that attached drawing carries out the dead angle determination unit 103 formed in this way
It is bright.
Figure 10 is to represent the definition graph of an example of dead angle determination processing that the dead angle determination unit 103 of embodiment 1 carries out.This
Outside, pair label identical with the same element labels of Fig. 4 omits detailed description.
Image 50 shown in Fig. 10m, image 50m+1And image 50m+2It is on time to the front of the vehicle during traveling
Between Sequentially continuous at the time of tm, moment tm+1And moment tm+2The image taken.In addition, in image 50m, image 50m+1And
Image 50m+2Appearing before one's eyes respectively has curved mirror 51m, curved mirror 51m+1And curved mirror 51m+2, in curved mirror 51m, curved mirror 51m+1With
And curved mirror 51m+2Further appear before one's eyes someone 60mAnd people 60m+1。
In this case, in continuous image 50 in temporal sequencemAnd image 50m+1Curved mirror 51mAnd curved mirror
51m+1 appears before one's eyes the people 60m and people 60 that identified portion 13 recognizesm+1.On the other hand, in image 50m+1Image later
50m+2Curved mirror 51m+2Do not appear before one's eyes out people, unidentified to people in identification part 13.Therefore, dead angle determination unit 103 is by image 50m+2
It is determined as being dead angle image.The reasons why can determine dead angle image to dead angle determination unit 103 using Figure 11 illustrates.
Figure 11 is the definition graph at the dead angle of curved mirror.
In curved mirror, there is also visual angles, and there are dead angle, that is, dead zones.Such as the vehicle at intersection is shown in FIG. 11
2 temporarily cease during moving object 61 as the people for multiplying bicyclem+1It is moved to moving object 61m+2Position, enter
Example in the case of the dead zone of curved mirror 72.In this case, the driver of the vehicle 2 during temporarily ceasing can be true
Recognize the moving object 61 appeared before one's eyes in curved mirror 71m+1, on the other hand, moving object 61 can not be confirmed in curved mirror 72m+2。
But it even if does not appear before one's eyes out moving object 61 in curved mirror 72m+2, moving object 61m+2It is also existing.Therefore, dead angle judges
Portion 103 by being judged as described above, by image 50m+2It is determined as that the moving object as people is likely to be present in curved surface
Mirror 51m+2Dead angle dead angle image.
In addition, in the case where dead angle determination unit 103 determined dead angle image, output processing part 102, which is generated and exported, to be made
The vehicle control information that vehicle temporarily ceases, until directly or in curved mirror seeing artificial stop from vehicle.In addition, at dead angle
After determination unit 103 determined dead angle image, in the case where also not becoming to see people even across certain time, it is determined as
Time-out proceeds by dead angle determination processing from initially.It is due to thinking to be determined as time-out：Even across certain time
Do not become in the case of seeing people, be present in the moving objects such as the people of the dead zone of curved mirror and enter in family or shop etc.,
Until what time no matter wait will not become seeing people.
Further, at least two images continuous in temporal sequence that dead angle determination unit 103 can also be in multiple images
The different object of middle size be detected processing unit 12a detection (identification) to, than at least two images in time series rearward
Objects in images be not detected in the case that processing unit 12a detection (identification) arrives, by the spectral discrimination rearward for it is above-mentioned extremely
Angle image.Such as when being illustrated using Figure 10, in image 50m+1In people 60 in the curved mirror appeared before one's eyesm+1Size be than
Forward image 50 in time seriesmIn the curved mirror 51 appeared before one's eyesmInterior people 60mSize it is big.This means that people 60 is approaching
Intersection.Also, than image 50m+1Image 50 in time series rearwardm+2Curved mirror 51m+2In do not appear before one's eyes someone.
This means that people 60 enters the dead angle near intersection.Therefore, dead angle determination unit 103 is based in image 50m+1In appear before one's eyes
People 60 in curved mirrorm+1Size and image 50 forward in time seriesmIn the curved mirror 51 appeared before one's eyesmInterior people 60m's
The comparison result of size and image 50m+2In the presence or absence of people 60, can be by image 50m+2It is determined as being dead angle image.As a result,
In people 60 far from intersection and in the case of entering dead angle, the image that people 60 does not appear before one's eyes in curved mirror is not judged as extremely
Angle image, therefore, it is possible to inhibit to occur useless action that is despite safety but making vehicle stopping etc..
102 ＞ of ＜ output processing parts
The size for the object that output processing part 102 is calculated based on calculation section 101 generates the drive supporting information of vehicle, defeated
Go out generated drive supporting information.Drive supporting information is according to corresponding with continuous at least two image of time series respectively
The change information of the respective size of object generate, and exported.Here, drive supporting information is included in change information
Represent size it is widened in the case of for make vehicle deceleration information and for vehicle far from object direction on move
At least one of information.In addition, drive supporting information be included in change information represent size it is widened in the case of be used for
The information for travelling vehicle after object is far from vehicle.In addition, drive supporting information is either the control of the movement of vehicle
Command information processed or be prompted to vehicle occupant prompt message.
Figure 12 is the figure of an example that the function for the output processing part 102 for representing embodiment 1 is formed.
In the present embodiment, as shown in figure 12, output processing part 102 has risk prediction portion 1031, information generation unit
1032 and information output part 1033.
Size of the risk prediction portion 1031 based on the object calculated by calculation section 101, prediction vehicle hold state is unchangeably
The danger predicted during traveling.The danger either danger level or dangerous content (such as contact, collision,
Be involved in and (involve) etc.).Risk prediction portion 1031 according to with the ruler of the corresponding object of continuous at least two image of time series
Very little variable quantity judges danger level.Here, as an example, danger level can be predicted according to the variable quantity of the size of object
This case illustrates.
Figure 13 is the schematic diagram for the variable quantity for representing the dimension of object in Fig. 9 (b).In addition, pair similary with (b) of Fig. 9
Element mark identical label, omit detailed description.Schematically illustrate in fig. 13 as zoom in or out to by
The people 60 ' of object, the people 60 ' to appear before one's eyes in the curved mirror for the reference dimension that dimension of object calculation section 1012 calculatesn+1And people
60′n+2。
As shown in figure 13, risk prediction portion 1031 is from people 60 ', people 60 'n+1And people 60 'n+2Size become larger successively understand
The people is close to curved mirror 51 i.e. close to intersection.Also, risk prediction portion 1031 is in people 60 ', people 60 'n+1And
People 60 'n+2Size variable quantity it is big when, it is known that the people is big close to the speed and velocity variations of curved mirror 51, therefore,
It can be determined that danger level is high.It is, as people 60 ', people 60 'n+1And people 60 'n+2Size variable quantity be more than threshold value when,
It can be judged as that the people rapidly touches close to there are the intersection of curved mirror 51, with carrying the vehicle of information processing unit 10
The possibility hit or contacted is high.In this way, risk prediction portion 1031 can predict danger according to the variable quantity of the size of object
Degree.
The size for the object that information generation unit 1032 is calculated based on calculation section 101 generates the drive supporting information of vehicle.Letter
The drive supporting information of vehicle can also be generated according to the size of object by ceasing generating unit 1032.For example, information generation unit 1032
It can be according to the change information next life of the respective size of object corresponding with continuous at least two image in temporal sequence respectively
Into drive supporting information.In addition, information generation unit 1032 can also on the basis of the size of object the also state based on vehicle
To generate the drive supporting information of vehicle.
In addition, information generation unit 1032 can also be according to the variation under the situation advanced by vehicle towards the direction of curved mirror
The danger of information prediction generates drive supporting information.Information generation unit 1032 can also be based on pre- by risk prediction portion 1031
The danger measured generates vehicle control information.
In addition, information generation unit 1032 can also generate the information of danger level for representing to be predicted by risk prediction portion 1031
It is used as drive supporting information.For example, information generation unit 1032 can also generate expression by risk prediction portion 1031 according to by
The information of the danger level of the variable quantity judgement of the size of the corresponding object of continuous at least two image of time series, is used as and drives
Sail support information.
In addition, information generation unit 1032 can also the judgement based on the dead angle image of dead angle determination unit 103 as a result, generation make
The vehicle control information that vehicle temporarily ceases.Specifically, information generation unit 1032 can also be continuous more in temporal sequence
In at least one of a image image by detection process portion 12a detection (identification) to object than at least one image when
Between be not detected in image in sequence rearward in the case that processing unit 12a detections (identification) arrive, generation drive supporting information.Separately
Outside, size is different at least two images continuous in temporal sequence that information generation unit 1032 can also be in multiple images
Object be detected processing unit 12a detection (identification) to, in the objects in images than at least two images in time series rearward
In the case that detected processing unit 12a detections (identification) are arrived, drive supporting information is generated.
Information output part 1033 exports the drive supporting information generated by information generation unit 1032.
Hereinafter, it is illustrated using an example that the output that attached drawing carries out the output processing part 102 formed in this way is handled.
Figure 14 is the figure of an example for the curve turning road for representing embodiment 1.Figure 15 is the output for representing embodiment 1
The definition graph of an example of output processing that processing unit 102 carries out.
Curve turning road shown in Figure 14 is an example in the place for looking into the distance condition difference.It is shown in FIG. 14 equipped with this reality
Apply the situation that the information processing unit 10 of mode or the vehicle 2 of system 1 are advanced on curve turning road.In addition, in Figure 14
It is middle as object one to be illustrated vehicle 3, it shows from vehicle 2 and is able to confirm that the vehicle 3 appeared before one's eyes in curved mirror 73
Situation.
In this case, output processing part 102 is that risk prediction portion 1031 zooms in or out i.e. according to by calculation section 101
The change in size for the object appeared before one's eyes in the curved mirror of reference dimension is standardized as, judges danger level.
More specifically, as shown in figure 15, the object appeared before one's eyes in curved mirror of the risk prediction portion 1031 after standardization
Change in size is when becoming smaller or not having change in size, is determined as that danger level is low.Here, output processing part 102 is information generation
Portion 1032 can also generate the information of the low this case of danger level for representing to be judged by risk prediction portion 1031.In addition, such as Figure 15
Shown, information generation unit 1032 can also be based on the low this case of danger level, and generation represents the controls such as the speed of no vehicle 2
The vehicle control information of change.In example shown in Figure 14, vehicle of the risk prediction portion 1031 in the curved mirror 73 after standardization
3 change in size is when becoming smaller or not having change in size, it can be determined that vehicle 3 is on the direction far from vehicle 2
During traveling or during stopping, therefore, it is determined as that danger level is low.Also, information generation unit 1032 can also generate expression
The vehicle control information of the change of controls such as the low information of danger level judged or the speed for representing no vehicle 2.
In addition, as shown in figure 15, the size of object that risk prediction portion 1031 appears before one's eyes in the curved mirror after standardization by
When gradual change is big, in being determined as that danger level is.Here, information generation unit 1032 can be both exported in representing that judged danger level is
Information, as shown in figure 15, can also be based on danger level be middle this case, generate to believe the vehicle control of vehicle deceleration
Breath.In the example shown in Figure 14, the change in size phase of vehicle 3 of the risk prediction portion 1031 in the curved mirror 73 after standardization
Than when threshold value a becomes larger, it can be determined that vehicle 3 is in during advancing close to the direction of vehicle 2, therefore,
In being determined as that danger level is.Also, information generation unit 1032 can also generate represent judged danger level be in information or
Person is used for the vehicle control information that vehicle 2 is made to slow down.
In addition, as shown in figure 15, the size for the object that risk prediction portion 1031 appears before one's eyes in the curved mirror after standardization is anxious
When drastic change is big, it is determined as danger level height.Here, information generation unit 1032 can also generate the high letter of the danger level for representing judged
Breath.In addition, as shown in figure 15, information generation unit 1032 can also be based on the high this case of danger level, generate to subtract vehicle
Speed and the vehicle control information moved on the direction far from object.In the example shown in Figure 14, risk prediction portion 1031
When the change in size of the vehicle 3 in curved mirror 73 after standardization is bigger than the threshold value b for being more than threshold value a, it can be determined that vehicle 3
During being advanced close on the direction of vehicle 2 with big speed, therefore, it is determined as danger level height.Also,
Information generation unit 1032 can also generate the high information of the danger level for representing judged or for vehicle 2 to be made to slow down and remote
Side from vehicle 3 is upwardly so that even if vehicle 3 is also at the position that will not be collided beyond the track of curve turning road
Vehicle control information.
In addition, in Figure 14 and Figure 15, vehicle 2 is on curve turning road during being travelled during advancing
Situation is illustrated, but not limited to this.It could also say that in the case where vehicle 2 is travelled in the intersection of intersection same
Sample.
Figure 16 is the figure of an example for the T words intersection for representing embodiment 1.Figure 17 is the output for representing embodiment 1
The definition graph of other an examples of output processing that processing unit 102 carries out.
T words intersection shown in Figure 16 is an example in the place for looking into the distance condition difference.It is shown in FIG. 16 equipped with this reality
Apply the situation that the information processing unit 10 of mode or the vehicle 2 of system 1 are just being travelled in T words intersection.In addition, in figure 16
One as object is illustrated vehicle 3, shows a case that be able to confirm that the vehicle 3 appeared before one's eyes in curved mirror 74 from vehicle 2.
In this case, as shown in figure 17, output processing part 102 is curved mirror of the risk prediction portion 1031 after standardization
In the change in size of object appeared before one's eyes when becoming smaller or there is no change in size, to be determined as that danger level is low.Here, output processing part
102 i.e. information generation unit 1032 can also generate the information for representing that the danger level judged by risk prediction portion 1031 is low.In addition, such as
Shown in Figure 17, information generation unit 1032 can also be based on the low this case of danger level, generate that vehicle 2 is made to temporarily cease, so
It is allowed to the vehicle control information to start to walk afterwards.In the example shown in Figure 16, curved mirror 74 of the risk prediction portion 1031 after standardization
In the change in size of vehicle 3 appeared before one's eyes for when becoming smaller or there is no change in size, it can be determined that vehicle 3 is in far from vehicle
During 2 side travels upwardly or during stopping, therefore, it is determined as that danger level is low.Also, information generation unit 1032 can also
The judged low information of danger level of generation expression or the vehicle control started to walk for vehicle 2 to be made to temporarily cease and then be allowed to
Information.
In addition, as shown in figure 17, the size for the object that risk prediction portion 1031 appears before one's eyes in the curved mirror after standardization becomes
It turns to when becoming larger, is determined as danger level height.Sentenced here, information generation unit 1032 can also generate expression by risk prediction portion 1031
The high information of fixed danger level.In addition, as shown in figure 17, information generation unit 1032 can also be based on the high this case of danger level,
It generates to temporarily cease vehicle 2 and then confirming the vehicle 2 for being used as object by sensor etc. by being allowed to later
The vehicle control information of starting.In the example shown in Figure 16, vehicle of the risk prediction portion 1031 in the curved mirror 74 after standardization
3 change in size is when becoming larger, it can be determined that vehicle 3 is in during advancing towards the direction of vehicle 2, therefore, sentences
It is set to danger level height.Also, information generation unit 1032 can also generate the high information of the danger level for representing judged or be used for
It temporarily ceases vehicle 2, vehicle 2 is being confirmed by being allowed to the vehicle control information to start to walk later by sensor etc..
In addition, in above-mentioned, the change information to size is that the example of the variable quantity of size is illustrated, but size
Change information can also be the information for the changing content for representing size.For example, the changing content of size can also be become large-sized
(or becoming smaller), size become more than preliminary dimension and (become smaller than scheduled size) etc..
It is that the example of control instruction information is illustrated, but drive supporting to drive supporting information in addition, in above-mentioned
Information can also be prompt message.For example, prompt message can also be the aftermentioned information for representing danger, represent to driver
Recommendation operation information.
[work of system 1]
Then, the information processing method of system 1 formed as described above is illustrated.Figure 18 is to represent embodiment 1
System 1 information processing method summary flow chart.Figure 19 is the information processing method for the system 1 for representing embodiment 1
Details flow chart.In addition, marking identical label to same element in Figure 18 and Figure 19, omit detailed
Explanation.
First, as shown in figure 18, system 1 is reflected based on image information to the curved mirror appeared before one's eyes in the picture and in curved mirror
Existing object is detected (S10).Then, system 1 calculates the size for the curved mirror appeared before one's eyes in the picture and the difference of reference dimension
It is different, calculate the size for the object appeared before one's eyes in curved mirror (S11) using the difference.Then, system 1 is based on calculating in S11
Object size, generate the drive supporting information (S12) of vehicle.Also, system 1 exports the drive supporting generated in S12
Information (S13).
More specifically, as shown in figure 19, first, system 1 carries out image acquirement processing, in image acquirement processing,
The figure of image for representing that the shooting of the filming apparatus by being equipped on the vehicle during advancing obtains is obtained by image acquiring section 11
As information (S101).
Then, system 1 carries out the processing using Figure 18 S10 illustrated.More specifically, system 1 carries out pair in S10
The detection process (S102) that the curved mirror appeared before one's eyes in the image obtained in S101 is detected.Then, system 1 carry out to
The identifying processing (S103) of (detection) is identified in the object in curved mirror appeared before one's eyes in the image obtained in S101.
Then, system 1 carries out the processing using Figure 18 S11 illustrated.More specifically, in S11, system 1 carries out
Calculate the difference calculating processing (S104) of the size of the curved mirror detected in S102 and the difference of reference dimension.Then, it is
System 1 is used the difference calculated in S104 to calculate the object for the object in the curved mirror recognized in S103
Size dimension of object calculating processing (S105).Further, since for the detailed of the processing that is carried out in S104 and S105
Situation is as described above, therefore, to omit detailed description in this.
Then, system 1 carries out following dead angle determination processing：In continuous three images in temporal sequence temporally
Do not known in S103 in the image after two images by the object that S103 is recognized in two images of Sequentially continuous
In the case of being clipped to, the spectral discrimination after this is likely to be present in the dead angle image at the dead angle of curved mirror for the object
(S106)。
Here, system 1 determines whether that recognizing the object in S103 after dead angle image is judged by S106 occurs
Or it have passed through certain time (S107) after the judgement.System 1 is after dead angle image is judged by S106 in S103
When there is the object or time-out i.e. after by S106 judgement dead angle images even across certain time also in S103
In in the case of the unidentified appearance to the object ("Yes" in S107) enter processing (S12) then.Situation other than herein
Under ("No" in S107), the determination processing of S107 is repeated.
Then, system 1 carries out the processing using Figure 18 S12 illustrated.More specifically, in S12, system 1 carries out
Following risk prediction processing (S108)：Based on the size of the object calculated in S105, predict constant in vehicle hold state
And the danger level judged when travelling.Then, system 1 carries out following information generation processing (S109)：Generation represents
The information of the danger level judged in S108 is used as drive supporting information and/or based on the danger level generation judged in S108
For controlling the vehicle control information of vehicle.Further, since for the details of the processing carried out in S108 and S109
It is as described above, therefore, to omit explanation in this.
Then, system 1 carries out the processing using Figure 18 S13 illustrated.More specifically, in S13, system 1 carries out
Export the information output processing (S110) of the drive supporting information generated in S109 etc..
In addition, the processing sequence of S103, S104, S105 are not limited to the situation shown in Figure 19.It can also be in the laggard of S104
Row S103, S105 is carried out.In addition, the processing of S106 and S107 can also carry out before the calculating of S11 processing.
[effect of embodiment 1 etc.]
As previously discussed, 1 information processing unit 10 or system 1 according to embodiment, by being looked into the distance in intersection etc.
The place of prestige condition difference utilizes curved mirror, can determine danger level, can be generated based on the danger level judged and export vehicle
Drive supporting information.Safe driving thereby, it is possible to the vehicle to being equipped with information processing unit 10 is supported.
Specifically, in the place for looking into the distance the intersection of condition difference etc., the object of personage to appear before one's eyes in curved mirror etc.
In the case that the amount of movement of body is big, the possibility which occurs (running out of) suddenly to place is high.In this case, it needs to make vehicle
Ahead of time slow down to avoid danger.
Then, in the present embodiment, due to the curved surface appeared before one's eyes in the image that can be obtained from the vehicle during traveling
The size of mirror can change, and therefore, the standardization of scheduled reference dimension is consistently adjusted into the size for being about to curved mirror.By
The amount of movement of the object of personage in curved mirror etc., can be transformed to the size i.e. size of the object in curved mirror by this,
Therefore, it is possible to judge danger level according to the variable quantity of size.In this way, the information processing unit 10 of embodiment 1 by using
Curved mirror can determine danger level.
In addition, in the case that the vehicle in such as automatic Pilot has information processing unit 10, which can be as above
It is described it is such judge danger level using curved mirror, therefore, it is possible to carry out vehicle control according to the danger level judged.
In addition, as described above, or, drive supporting information according to continuous at least two image in temporal sequence
The change information of the corresponding respective size of object generates, and change information represents the expansion of size.In this case, it drives
Sailing support information both can be including the information for making vehicle deceleration and for making vehicle be moved on the direction far from object
At least one of information, can also include in object far from vehicle after travel vehicle information.This be because
According to these can be ensured that the safety close to the object of intersection and vehicle.
In addition, drive supporting information can also also be generated on the size basis of object based on the state of vehicle.This is
Because can be taken place without according to the state (transport condition, halted state etc.) of vehicle be useless work control.
In addition, drive supporting information can also be generated according to the size of object.This is because even if without using
The processing of the change information of size can also predict danger.
As previously discussed, 1 information processing unit 10 or system 1 according to embodiment, by being looked into the distance in intersection etc.
The place of prestige condition difference utilizes curved mirror, and the safe driving of vehicle can be supported.
(variation 1)
In the embodiment 1, danger is judged to the variable quantity of the size according to the object in the curved mirror after standardization
The situation of degree is illustrated, but not limited to this.Or identification part 13 identifies the category for the object appeared before one's eyes in curved mirror
Property, calculation section 101 considers the attribute to judge danger level.In this variation, to further considering with appearing before one's eyes in curved mirror
Object possessed by movement speed related attribute judge that the situation of danger level illustrates.
Figure 20 and Figure 21 is the risk prediction processing that the output processing part 102 for the variation 1 for representing embodiment 1 carries out
An example definition graph.In addition, pair describing identical description, detailed description will be omitted with the same contents of Figure 15 and Figure 17.
The risk prediction processing that the output processing part 102 of the variation 1 at curve turning road carries out is shown in FIG. 20
An example, be shown in FIG. 21 the variation 1 at T words intersection output processing part 102 carry out risk prediction processing
An example.In addition, based on it is low, in or high danger level vehicle control information can be such as Figure 15 and as shown in Figure 17, because
This, the illustration is omitted in Figure 20 and Figure 21.
Object is in the case of personage in such as Figure 20 and as shown in Figure 21, risk prediction portion 1031 and Figure 15 and figure
Danger level is similarly judged when 17.On the other hand, it is bicycle, motorcycle or the vapour faster than the movement speed of personage in object
In the case of vehicle, risk prediction portion 1031 is determined as the danger level than object for personage Shi Gao according to movement speed.
As previously discussed, in this variation, output processing part 102 is according to the attribute of the object recognized by identification part 13
To generate drive supporting information, the generated drive supporting information of output.
(variation 2)
In variation 1, the related attribute of the movement speed that has to the object for also considering with appearing before one's eyes in curved mirror comes
The situation of judgement danger level is illustrated, but not limited to this.It is dangerous pre- in the case that the object appeared before one's eyes in curved mirror is people
Survey portion 1031 can also also judge the attribute related with the age of the people to judge danger level.Below using the situation as variation
2 illustrate.
Figure 22 and Figure 23 is the risk prediction processing that the output processing part 102 for the variation 2 for representing embodiment 1 carries out
An example definition graph.In addition, pair describing identical description with the same contents of Figure 15 and Figure 17, omit specifically
It is bright.
Figure 22 represents the risk prediction processing that the output processing part 102 in variation 2 at curve turning road carries out
An example, Figure 23 represent an example of risk prediction processing that the output processing part 102 in variation 1 at T words intersection carries out.
In addition, based on it is low, in or high danger level vehicle control information may be such as Figure 15 and as shown in Figure 17, therefore,
Diagram is also omited in Figure 22 and Figure 23.
As shown in Figure 22 and Figure 23, in the case where object is personage and the personage is child or old man, risk prediction
Portion 1031 is determined as and the same danger levels of Figure 15 and Figure 17.On the other hand, it is personage in object and the personage is child
In the case of other people other than son and old man, it is determined as than personage high danger level when being child or old man.
As previously discussed, in this variation, output processing part 102 is in the attribute of the object recognized by identification part 13
In the case of personage, drive supporting information change is made according to the information related with the age of the object identified by identification part 13
And export the drive supporting information.In addition, in above-mentioned, pair information related with the age is that the example of the generation of personage carries out
Illustrate, but the information related with the age can also be the age or age of personage.
(variation 3)
In variation 2 to the object appeared before one's eyes in curved mirror be people in the case of also consider it is related with the age of the people
Attribute judge that the situation of danger level is illustrated, but not limited to this.The object appeared before one's eyes in curved mirror is the situation of people
Under, risk prediction portion 1031 can also further also consider the people whether do not paid attention to action (attention is insufficiently gone
It is dynamic) this attribute judges danger level.Specifically, do not notice that action is non-to see front action.For example, non-see front action packet
It includes and watches the portable terminals such as smart phone or books attentively.It is illustrated below using the situation as variation 3.
Figure 24 and Figure 25 is the risk prediction processing that the output processing part 102 for the variation 3 for representing embodiment 1 carries out
An example definition graph.In addition, pair describing identical statement with the same contents of Figure 15 and Figure 17, omit specifically
It is bright.
It is shown in FIG. 24 at the risk prediction that the output processing part 102 in variation 3 at curve turning road carries out
An example of reason is shown in FIG. 25 at the risk prediction that the output processing part 102 in variation 3 at T words intersection carries out
An example of reason.In addition.Based on it is low, in or high danger level vehicle control information can be as shown in Figure 15 and Figure 17 that
Therefore sample, also omits diagram in Figure 24 and Figure 25.
As shown in Figure 24 and Figure 25, object be personage and the personage is not carrying out looking at portable terminal and into
The mobile action of row it is non-see that the front action i.e. personage is not walkinged on one side look at portable terminal while in the case of, risk prediction
Same danger level when being determined as with Figure 15 and Figure 17 of portion 1031.On the other hand, object be personage and the personage just
Carry out it is non-see that the front action i.e. personage is walkinged on one side look at portable terminal while in the case of, risk prediction portion 1031
It is judged to not carrying out than the personage non-seeing danger level high during the action of front.In addition, in above-mentioned, illustrate not pay attention to
Action is the non-example for seeing front action, although not noticing that action can also looked at the front of personage but looking at
The row of the specific object of the action of side on the upper side or partial below or the perambulator or the ball that are just look at the front for being located at personage etc.
It is dynamic.
In this way, in this variation, output processing part 102 is personage in the attribute of the object recognized by identification part 13
In the case of, recognize whether object is not paid attention to taking action to generate and export drive supporting letter according to by identification part 13
Breath.
(embodiment 2)
It in the embodiment 1, can be by using curved mirror to row in the place for looking into the distance condition difference of intersection etc.
Information processing unit supported into the safe driving of the vehicle of period etc. is illustrated, but not limited to this.Stop temporarily
In vehicle in only, safe driving can also be supported using curved mirror.Hereinafter, it is carried out the situation as embodiment 2
Explanation.
[composition of system 1A]
Figure 26 is the block diagram of an example of the composition for the system 1A for representing embodiment 2.In addition, pair with the same elements of Fig. 1
Identical label is marked, omits detailed description.
For system 1A shown in Figure 26 compared with the system 1 of embodiment 1, the composition of information processing unit 10A is different.System
1A can also be equipped on the vehicle such as automobile in the same manner as system 1, can export the vehicle by using curved mirror
Drive supporting information.
In the present embodiment, in order to support the safe driving of the vehicle in temporarily ceasing, information processing unit
10A is also contemplated for the position of the object in curved mirror.In addition, image acquiring section 11 and detection process portion 12A can also be with embodiment party
Formula 1 is similarly formed in information processing unit 10A.
[information processing unit 10A]
Information processing unit 10A in embodiment 2 exports the driving branch of the vehicle in temporarily ceasing by curved mirror
Help information.In the present embodiment, in order to support the safe driving of the vehicle in temporarily ceasing, information processing unit
10A is also contemplated for the position of the object in curved mirror.For its reason, illustrated using Figure 27 A~Figure 27 C.Figure 27 A are to represent
An example of object 63 appeared before one's eyes in the curved mirror 75 that the Driver Vision of vehicle 2a temporarily ceased in intersection recognizes
Figure.Figure 27 B and Figure 27 C are the definition graphs of the characteristic of curved mirror.
Such as the example being shown below in Figure 27 A：The driver of vehicle 2A temporarily ceased in intersection
In the case of visual identity to the curved mirror 75 for being arranged on the intersection, as the object 63 appeared before one's eyes in curved mirror 75, energy
Enough confirm the people of cycling.In Figure 27 A, since the top in the inside, that is, curved mirror in curved mirror 75 is appeared before one's eyes out work
For the object 63 of people, the driver of vehicle 2A can regard object 63 as and be present in road from the point of view of vehicle 2a like that as shown in figure 27b
Inboard position.But due to the characteristic of curved mirror that left and right is appeared before one's eyes on the contrary, in fact, object 63 is present in such as Figure 27 C
Shown position is i.e. from the position of the front side of the road from the point of view of vehicle 2a.
Therefore, the information processing unit 10A of present embodiment is on the basis of position for having also contemplated the object in curved mirror
On, the drive supporting information of output vehicle 2a.
More specifically, as shown in figure 26, the information processing unit 10A of present embodiment has calculation section 101A, output
Processing unit 102A and dead angle determination unit 103.Information processing unit 10A and the information processing apparatus of embodiment 1 shown in Figure 26
It puts 10 to compare, the composition of calculation section 101A and output processing part 102A is different.Hereinafter, to calculation section 101A and output processing part
The details of the composition of 102A etc. illustrates.
＜ calculation section 101A ＞
Figure 28 is the figure of an example that the function for the calculation section 101A for representing embodiment 2 is formed.Figure 29 A are to represent Figure 28 institutes
The figure of an example of the position for the object that the object space calculation section 1013 shown calculates.Figure 29 B are the road surface areas represented shown in Figure 28
The figure of an example in the road surface region that domain calculation section 1015 calculates.Figure 30 is during the central shaft calculation section 1016 shown in Figure 28 calculates
The definition graph of the calculation method of mandrel.
The calculation section 101A of embodiment 2 calculates the position of the object in the curved mirror detected by detection process portion 12A.
More specifically, calculation section 101A is calculated in the curved mirror detected by test section 12 and the filming apparatus by being equipped on vehicle
The obtained image of shooting in the position of the object that is recognized by identification part 13 in the curved mirror appeared before one's eyes in curved mirror.
For example, as shown in figure 28, calculation section 101A have object space calculation section 1013, dimension of object calculation section 1014,
Road surface region calculation section 1015 and central shaft calculation section 1016.In addition, calculation section 101A only can also have object space to calculate
Go out portion 1013 or only there is object space calculation section 1013 and dimension of object calculation section 1014.
Object space calculation section 1013 calculates the position of the object in curved mirror.More specifically, object space calculation section
1013 calculate the curved mirror that is detected by test section 12 and continuous with time series by the front to the vehicle in temporarily ceasing
Position of the object recognized in the curved mirror appeared before one's eyes in image obtained from shooting by identification part 13 in curved mirror.For example,
As shown in figure 29 a, object space calculation section 1013 calculates the object 63 as the people that ride bicycle in the region of curved mirror 75
Interior position.In addition, in the case where calculation section 101A only has object space calculation section 1013, object space calculation section 1013
The position that the object in curved mirror can also be calculated is kept right or is leaned on compared to the threshold value for representing the precalculated position in curved mirror
It is left.In addition, object space calculation section 1013 can also calculate the position of the object in curved mirror compared in expression curved mirror
The threshold value in precalculated position is top or on the lower.
Dimension of object calculation section 1014 calculates the size of the object on the basis of curved mirror.More specifically, dimension of object
Calculation section 1014 calculates the curved mirror that is detected by test section 12 and by the front to the vehicle in temporarily ceasing with time sequence
The size of object recognized in the curved mirror appeared before one's eyes in image obtained from row are continuously shot by identification part 13.
Road surface region calculation section 1015 calculates the road surface region in curved mirror.More specifically, road surface region calculation section
1015 calculate as the road surface region in the region of road appeared before one's eyes in the curved mirror for representing to be detected by test section 12.For example, such as
Shown in Figure 29 B, road surface region calculation section 1015 calculates the road surface region for the road for representing existing in the region of curved mirror 75
80.Road surface region calculation section 1015 can in advance be used by using deep learning (Deep Learning) of convolutional layer etc.
The image of a large amount of curved mirror of road of appearing before one's eyes is learnt, and thus calculates road surface region as described above.
Central shaft calculation section 1016 calculates the central shaft in road surface region calculated by road surface region calculation section 1015.More specifically
For, such as shown in figure 30, central shaft calculation section 1016 in the road surface region 80 calculated by road surface region calculation section 1015,
First the straight line for the curb for being equivalent to road surface (being known as curb straight line) is calculated using Hough transform.Then, central shaft calculation section
1016 calculate the position by curb straight line identical the distance D started from the endpoint of the curb straight line calculated and and curb
Line L1, the line L2 of line orthogonal.Similarly, it calculates through the identical road of the distance started from the endpoint of the curb straight line calculated
The position of shoulder straight line and with the line L3 and line L4 and line L5 and line L6 of curb line orthogonal.Also, central shaft calculation section 1016
The straight line for calculating the intersection point for linking these lines is used as the central shaft 82 in road surface region.
In addition, as described above, to central shaft calculation section 1016 by calculating line L1 and line L2, line L3 and line L4 and line
This three groups of lines of L5 and line L6 are illustrated to calculate the situation of central shaft 82, but not limited to this.Central shaft 82 can it is minimum by
The intersection point of two is obtained, and therefore, central shaft calculation section 1016 can also be from the intersection point and curb of the line L1 calculated and line L2
The endpoint of straight line calculates.In addition, central shaft calculation section 1016 can also calculate four groups or more of line, least square method is used
(least squares method), the straight line for calculating the intersection point for the line for making four groups or more and the square error minimum of straight line are used as central shaft
82。
＜ output processing part 102A ＞
Output processing part 102A drives branch based on the position of the object in the curved mirror calculated by calculation section 101A to generate
Information is helped, and exports generated drive supporting information.Here, drive supporting information can both include the control of the movement of vehicle
Command information can also include being prompted to the prompt message of the occupant of vehicle.
Figure 31 is the illustration that the function for the output processing part 102A for representing embodiment 2 is formed.Figure 32 is to represent to implement
The figure of an example of the position of object in the curved mirror of mode 2.
In the present embodiment, as shown in figure 31, output processing part 102A has risk prediction portion 1031A, information generation
Portion 1032A and information output part 1033A.
《Risk prediction portion 1031A》
Risk prediction portion 1031A predicts danger based on the position of the object in the curved mirror calculated by calculation section 101A
Property.For example, risk prediction portion 1031A both can vehicle under the situation advanced towards the direction of curved mirror from the curved surface of object
Position prediction in mirror is dangerous, can also be according to object and the region in the curved mirror determined by the road appeared before one's eyes in the picture
Between position relationship predict danger.
In the present embodiment, risk prediction portion 1031A based on the position calculated by object space calculation section 1013 and by
The central shaft that central shaft calculation section 1016 calculates, prediction are dangerous.Specifically, the danger is either danger level, also may be used
To be dangerous content (such as contact, collide, be involved in).
Risk prediction portion 1031A by the position that object space calculation section 1013 calculates compared to by central shaft calculation section
In the case that 1016 central shafts calculated are kept right, high danger level when being judged to keeping left compared to the central shaft than its position.Example
As shown in figure 32, if the lower end of object 63 compared to road surface region central shaft 82 and positioned at keeping right or top position,
Object 63 can be then defined as and be present in the inside in the region of curved mirror 75.Also, in object 63 in the region of curved mirror 75
Inside be present in the inside when, due to object 63 from the point of view of vehicle positioned at the position of the front side of road, so in temporarily ceasing
In the presence of the danger collided with object 63 during vehicle start.Therefore, risk prediction portion 1031A in object 63 in the region of curved mirror 75
When being inside present in the inside, danger level is determined as height.On the other hand, if the lower end of object 63 is located at and leans on compared to central shaft 82
Left or position on the lower can then be defined as object 63 and be present in front in the region of curved mirror.Also, exist in object 63
When in front of being present in the region of curved mirror 75, object 63, positioned at the inboard position of road, therefore, temporarily stops from the point of view of vehicle
Even if the vehicle starting in only, the danger collided with object 63 are also low.Therefore, risk prediction portion 1031A can exist in object 63
Danger level is determined as when in front of being present in the region of curved mirror 75 low.
Hereinafter, more specific description is carried out to the situation using attached drawing.
Figure 33, Figure 34 and Figure 35 are the danger in the case that the object 64 in embodiment 2 is located at the inboard of curved mirror 75
The definition graph of dangerous degree.It is shown in FIG. 33 from the vehicle 2a acquirement objects 64 temporarily ceased and appears before one's eyes in the figure of inboard curved mirror 75
The situation of picture.Position and its size for the object 64 for being present in curved mirror 75 in (a) of Figure 34 are shown, belonged at (b) of Figure 34
In the enlarged drawing of the curved mirror 75 of (a) of Figure 34, the relationship of object 64 and road is shown.The object shown in Figure 34 is shown in FIG. 35
The position relationship of body 64, vehicle 2a and curved mirror 75.
As shown in figure 35, the object 64 as people is set as from the point of view of vehicle 2a in the side of the lateral intersection in the front of road
To walking.Also, t at the time of the system 1A for being set as being equipped on the vehicle 2a temporarily ceased obtains as shown in figure 34j, moment tj+1
Or moment tj+2Image 50j, image 50j+1Or image 50j+2.In this case, system 1A, that is, information processing unit 10A
Calculate object 64j, object 64j+1Or object 64j+2Position in curved mirror 75.In image 50j, image 50j+1Or image
50j+2In, object 64j, object 64j+1Or 64j+2It is present in inboard, therefore, object 64j、64j+1Or 64j+2Come from vehicle 2A
It sees positioned at the position of the front side of road.It that is to say, in the presence of the danger collided with object 64 when the vehicle 2a in temporarily ceasing starts to walk
Danger, therefore, is determined as height by danger level.
In addition, the system 1A for being equipped on the vehicle 2a temporarily ceased can also obtain image 50j、50j+1And 50j+2, calculate
Go out object 64j、64j+1Or 64j+2Position and its size.In this case, according to object 64j、64j+1Or 64j+2Size
Become larger this continuity, it is known that object 64 is approaching the vehicle 2a in temporarily ceasing at intersection.Therefore, also may be used
With with object 64j+1It will be for object 64 etc. comparingj+2Danger level be determined as higher.In this way, risk prediction portion 1031A also may be used
Danger level is judged with size changing amount according to the object corresponding to continuous at least two image in temporal sequence.Namely
It is that risk prediction portion 1031A can also be in the ruler that the object on the basis of curved mirror has been calculated by dimension of object calculation section 1014
When very little, based on position of the object calculated by object space calculation section 1013 in curved mirror and by dimension of object calculation section
The size of 1014 objects calculated predicts danger level.
In addition, achieve image 50 in the system 1A for being equipped on the vehicle 2a temporarily ceasedj+2, image 50j+3In the case of,
In image 50j+3Curved mirror 75 in be not present object 64.But t at the time of acquirement before continuity is justj+2Image
50j+2Curved mirror 75 in there are objects 64j+2, therefore, by image 50j+3It is determined as that the object 64 is likely to be present in curved mirror 75
Dead angle dead angle image, with object 64j+2It will be for image 50 etc. comparingj+3Danger level be determined as higher.
Figure 36, Figure 37 and Figure 38 be in embodiment 2 in the case where the front side of curved mirror 75 is there are object 64
Danger level definition graph.It is shown in FIG. 36 from the vehicle 2a acquirement objects 64 temporarily ceased and appears before one's eyes in the curved mirror of front side
The situation of 75 image.In the position for the object 64 that (a) of Figure 37 shows to be present in curved mirror 75 and its size, (b) of Figure 37
Belong to the enlarged drawing of the curved mirror 75 of (a) of Figure 37, the relationship between object 64 and road is shown.Figure 37 institutes are shown in FIG. 38
The position relationship of object 64, vehicle 2a and curved mirror 75 shown.
As shown in figure 38, the object 64 as people is set as from the point of view of vehicle 2a in the inboard direction to intersection of road
Walking.Also, t at the time of the system 1A for being set as being equipped on the vehicle 2a temporarily ceased obtains as shown in figure 37k, moment tk+1Or
Person moment tk+2Image 50k, image 50k+1Or image 50k+2.In system 1A, that is, information processing unit 10A calculatings
Object 64k, object 64k+1Or object 64k+2Position in curved mirror 75.In image 50k, image 50k+1Or image 50k+2
In, object 64k, object 64k+1Or object 64k+2It is present in front side, therefore, object 64k, object 64k+1Or object 64k+2
Positioned at the inboard of road from the point of view of vehicle 2a.It that is to say, the danger collided in the vehicle 2a startings in temporarily ceasing with object 64
Danger is low, therefore, danger level is determined as low.
In addition, the system 1A for being equipped on the vehicle 2a temporarily ceased can also obtain image 50k, image 50k+1And image
50k+2, calculate object 64k, object 64k+1Or object 64k+2Position and its size.In this case, according to object 64k, object
Body 64k+1Or object 64k+2Size become larger this continuity, it is known that object 64 is approaching temporarily at intersection
Vehicle 2a in stopping.It accordingly it is also possible to will be for object 64k+2Danger level be determined as than object 64k+1It is contour.In this way, danger
Dangerous prediction section 1031A can also according to continuous at least two image in temporal sequence respectively in the respective size of object appeared before one's eyes
Change information and the object position in curved mirror judge danger level.It that is to say, risk prediction portion 1031A can also basis
The position of object in the curved mirror calculated by calculation section 101A and the size of object that is calculated predict danger.
In above-mentioned, the change information of size is illustrated, but the variation of size for the example of the variable quantity of size
Information can also be the information for the changing content for representing size.For example, the changing content of size can also be become large-sized (or
Person becomes smaller), size becomes more than preliminary dimension and (becomes smaller than scheduled size) etc..
In addition, achieve image 50 in the system 1A for being equipped on the vehicle 2a temporarily ceasedk+2, image 50k+3In the case of,
In image 50k+3Curved mirror 75 in be not present object 64.But t at the time of due to being obtained before continuity is justk+2's
Image 50k+2Curved mirror 75 in there are objects 64k+2, so by image 50k+3It is determined as that the object 64 is likely to be present in curved mirror
The dead angle image at 75 dead angle, will be for object 64k+3Danger level be predicted as than object 64k+2Etc. highers.
In addition, risk prediction portion 1031A can also be based on the object calculated by calculation section 101A in curved mirror position,
The danger level predicted when constant to vehicle hold state judges.For example, risk prediction portion 1031A may be, by
Calculation section 101A calculate object position compared to curved mirror center line on the lower in the case of, be judged to comparing than the position
The high danger level when center line of curved mirror is top.In addition, risk prediction portion 1031A may be, by calculation section 101A
In the case that the position of the object of calculating is located at the right side compared to the precalculated position of curved mirror, it is determined as than the position compared to song
The precalculated position of face mirror and high danger level during positioned at left side.In addition, precalculated position is either the position of central shaft, it can also
It is predetermined the position of immediate vicinity.In addition, center line is, for example, the line by the center of curved mirror, including horizontal line,
Vertical line, oblique line, curve etc..It is horizontal in center line, danger level is judged according to up/down, at center
In the case that line is vertical line, danger level is judged according to right/left.In addition, in feelings of the center line for oblique line or curve
Under condition, danger level is judged according to one side/other party.
《Information generation unit 1032A》
Information generation unit 1032A drives branch based on the position of the object in the curved mirror calculated by calculation section 101A to generate
Help information.For example, information generation unit 1032A according to object and the region in the curved mirror that is determined by the road appeared before one's eyes in image it
Between position relationship generate drive supporting information.In addition, information generation unit 1032A can both be based on being calculated by calculation section 101A
Position of the object in curved mirror and drive supporting information is generated by the size of the calculation section 101A objects calculated, such as
Can according to continuous at least two image in temporal sequence respectively in the change information of the respective size of object appeared before one's eyes with
And position of the object in curved mirror generates drive supporting information.
More specifically, information generation unit 1032A can also be according to the object calculated by object space calculation section 1013
Position relationship between position and the central shaft calculated by central shaft calculation section 1016 generates drive supporting information.For example, letter
Generating unit 1032A is ceased in the case where object is located at the position top compared to central shaft in the picture, generates to make vehicle
Hide the drive supporting information of object.
In addition, information generation unit 1032A can also based on the object calculated by calculation section 101A position and curved mirror
Position relationship between center line generates drive supporting information.For example, information generation unit 1032A (i) in curved mirror from vehicle
From the point of view of towards object in the case of left be located at the position kept right than center line in the picture when or (ii) in curved mirror from vehicle
From the point of view of towards object in the case of right be located at the position to keep left than center line in the picture when, generate that vehicle is made to hide object
Drive supporting information.
In the present embodiment, information generation unit 1032A is generated according to the danger predicted by risk prediction portion 1031A
Drive supporting information.For example, information generation unit 1032A can also based on the danger predicted by risk prediction portion 1031A come
Generate vehicle control information.In addition, information generation unit 1032A can also advance according to vehicle to towards the direction of the curved mirror
Situation under, the danger of position prediction from the object in the curved mirror generate drive supporting information.
In addition, information generation unit 1032A can also export the letter of danger level for representing to be predicted by risk prediction portion 1031A
Breath is used as drive supporting information.For example, information generation unit 1032A output represent by risk prediction portion 1031 according to temporally
The information of the danger level of the variable quantity judgement of the size of the corresponding object of at least two images of Sequentially continuous, is used as driving branch
Help information.
Further, information generation unit 1032A can also the position based on the object calculated by object space calculation section 1013
This case that dead angle image is judged by dead angle determination unit 103, to generate the vehicle control letter including temporarily ceasing vehicle
The drive supporting information of breath.Specifically, information generation unit 1032A further can also be in continuous multiple figures in temporal sequence
As at least one of in image by the object that identification part 13 recognizes than at least one image in time series rearward
In the case that unrecognized portion 13 recognizes in image, drive supporting information is generated.In addition, information generation unit 1032A can also
Position based on the object appeared before one's eyes in an image and the unrecognized portion 13 of objects in images behind recognize this
Situation generates drive supporting information.
《Information output part 1033A》
The drive supporting information that information output part 1033A outputs are generated by information generation unit 1032A.
It is that the example of control instruction information is illustrated, but drive supporting to drive supporting information in addition, in above-mentioned
Information can also be prompt message.For example, prompt message can also be the information for representing aftermentioned danger, represent to driver
Recommendation operation information.
Here, it is illustrated using an example that attached drawing handles the output processing part 102A formed in this way the outputs carried out.
Figure 39 is the figure of an example for the T words intersection for representing embodiment 2.Figure 40 is the output for representing embodiment 2
The definition graph of other an examples of output processing that processing unit 102A is carried out.It is shown at the information equipped with present embodiment in Figure 40
The situation that the vehicle 2a of reason device 10A or system 1A is temporarily ceased at T words intersection.
In this case, as shown in figure 40, output processing part 102A, that is, risk prediction portion 1031A is in per time of object
When size becomes smaller or do not change, even if the position of the object in curved mirror 75 is inboard, also it is determined as that danger level is low.
This, output processing part 102A, that is, information generation unit 1032A can also export the danger for representing to be predicted by risk prediction portion 1031A
Spend low information.In addition, as shown in figure 40, information generation unit 1032A can also be based on the low this case of danger level, generation is used for
The vehicle control information that the vehicle 2a in temporarily ceasing is made to start to walk.In the example shown in Figure 39, risk prediction portion 1031A is in song
When the size of vehicle 3 in face mirror 75 becomes smaller or do not change, it is determined as that vehicle 3 is in the direction uplink far from vehicle 2a
During, therefore, it is determined as that danger level is low.Also, information generation unit 1032A can also generate the danger level for representing judged
Low information or the vehicle control information for the vehicle 2a in temporarily ceasing to be made to start to walk.
In addition, as shown in figure 40, the position of objects of the risk prediction portion 1031A in curved mirror 75 is front side and song
When becoming large-sized of object in face mirror 75, in being determined as that danger level is.Here, information generation unit 1032A can also generate table
Show the information in being by the risk prediction portion 1031A danger levels judged.In addition, as shown in figure 40, information generation unit 1032A also may be used
To be based on danger level as middle this case, generate that the vehicle 2a in temporarily ceasing is made to go slowly the vehicle control information of starting.
In the example shown in Figure 39, the position of vehicles 3 of the risk prediction portion 1031A in curved mirror 75 is front and curved mirror 75
When interior vehicle 3 becomes large-sized, during being determined as that vehicle 3 is in road laterally towards the traveling of the direction of vehicle 2a,
Therefore, in being determined as that danger level is.Also, information generation unit 1032A, which can also generate, represents that judged danger level is
Information or for the vehicle 2a in temporarily ceasing to be made to go slowly the vehicle control information of starting.
In addition, as shown in figure 40, the position of objects of the risk prediction portion 1031A in curved mirror 75 is inboard and curved surface
When becoming large-sized of object in mirror 75, is determined as danger level height.Here, information generation unit 1032A can also generate expression by
The danger level of risk prediction portion 1031A judgements is high information.In addition, as shown in figure 40, information generation unit 1032A can also base
Be high this case in danger level, generate for determined by sensor the vehicle 3 for being used as object by making later
The vehicle control information of vehicle 2a startings in temporarily ceasing.In the example shown in Figure 39, risk prediction portion 1031 is in curved surface
When becoming large-sized of vehicle 3 in mirror 75, it can be determined that vehicle 3 is in the front side of road in the direction towards vehicle 2a
During upper traveling, therefore, it is determined as danger level for height.Also, information generation unit 1032A, which can also be generated, represents what is judged
Danger level for high information or for confirmed by sensor vehicle 3 by the vehicle in making to temporarily cease later
The vehicle control information of 2a startings.
[action of system 1A]
Then, the information processing method of system 1A formed as described above is illustrated.Figure 41 is to represent embodiment 2
System 1A information processing method summary flow chart.Figure 42 is the information processing side for the system 1A for representing embodiment 2
The flow chart of the details of method.In addition, pair label identical with the same element labels of Figure 19, omits detailed description.
First, as shown in figure 41, system 1A is based on the image information curved mirror appeared before one's eyes in the picture of detection and in curved mirror
In the object (S20) appeared before one's eyes.Then, the position (S21) in the curved mirror that system 1A calculatings are detected in S20.Then, system 1A
Based on generating drive supporting information (S22) in the position that S21 is calculated.Also, the driving branch that system 1A outputs generate in S22
Help information (S23).
More specifically, as shown in figure 42, in the processing of S21, system 1A carries out following object space calculating processing
(S204)：Calculate the position of object in the curved mirror detected in S102, being recognized in S103.Then, system 1A
Carry out following dimension of object calculating processing (S205)：Calculate on the basis of the curved mirror detected in S102, in S103
In the size of object that recognizes.Then, system 1A carries out following road surface region calculating processing (S206)：It calculates in S102
Road surface region in the curved mirror detected.Then, system 1A carries out following central shaft calculating processing (S207)：It calculates
The central shaft in road surface region calculated in S206.In addition, the details about the processing carried out in S204~S207 is as above
It is described such, therefore, omit detailed description in this.
In addition, in the processing of the S22 illustrated in Figure 41, system 1A carries out following risk prediction processing (S221)：
Position based on the object calculated in S21 in curved mirror, judgement vehicle hold state is constant and danger that while advancing is predicted
Dangerous degree.Then, system 1A carries out following information generation processing (S222)：Generation represents the letter of danger level judged in S221
Breath is used as drive supporting information and/or generates the vehicle control for controlling vehicle based on the danger level judged in S221
Information processed.In addition, about the details of processing carried out in S221 and S222 as described above, therefore, to be omitted in
This explanation.
Also, the processing of S23 that system 1A progress illustrated in Figure 41.More specifically, in S23, system 1A into
The information output processing (S223) of the drive supporting information that row output generates in S222 etc..
[effect of embodiment 2 etc.]
As previously discussed, the 2 information processing unit 10A being related to or system 1A according to embodiment, in intersection etc.
The place of condition difference is looked into the distance, danger level can be judged by using curved mirror, can be generated based on the danger level judged
And export the drive supporting information of vehicle.Thereby, it is possible to the safe driving progress of the vehicle to being equipped with information processing unit 10A
It supports.
Specifically, look into the distance the place of condition difference in intersection etc., the object that has personage etc. of appearing before one's eyes in curved mirror
In the case of, which is possible to occur suddenly to the place.In this case, existing to work as makes the vehicle in temporarily ceasing rise in advance
The danger that can be collided during step with the object.In addition, there is curved mirror left and right to invert this characteristic of appearing before one's eyes.
Therefore, it is known that：It appears before one's eyes in the inboard of curved mirror in the case of the object for having personage etc., and in the front side of curved mirror
The situation of object of personage etc. of appearing before one's eyes is compared, the road of the near side of vehicle of the object in temporarily ceasing away from the place
It is nearby moved the curb side on road.It that is to say, appear before one's eyes in the inboard of curved mirror in the case of having the objects such as personage, and in curved surface
The appear before one's eyes situation of the objects such as personage of the front side of mirror is compared, and is touched with the object when the vehicle in making to temporarily cease is started to walk in advance
The danger hit is high.
Therefore, in the present embodiment, according to the song appeared before one's eyes in the obtained image of vehicle that can be from temporarily ceasing
The position of object in the mirror of face and its change in size whenever necessary judge danger level.In this way, the information processing of embodiment 2
Device 10A can judge danger level by using curved mirror.
In addition, in the case that the vehicle in such as automatic Pilot has information processing unit 10A, which can be as above
It is described it is such judge danger level using curved mirror, therefore, it is possible to carry out vehicle control according to the danger level predicted.
In addition, as described above, the information processing unit 10A of such as embodiment 2 can also in the picture be located in object
In the case of the position more top than central shaft, the drive supporting information for vehicle to be made to hide object is generated.Thereby, it is possible in object
Body hides object in the case of the vehicle, it can be ensured that the safety of object and vehicle.
In addition, the information processing unit 10A that such as embodiment 2 is related to may be that (i) is in curved mirror from the point of view of vehicle
When being located at the position kept right than center line in the picture towards object in the case of left or (ii) in curved mirror from the point of view of vehicle
When being located at the position to keep left than center line in the picture towards object in the case of right, generate that vehicle is made to hide driving for object
Sail support information.Object can be hidden in the case where object is close to vehicle as a result, it can be ensured that the safety of object and vehicle
Property.In addition, complicated processing as the central shaft for determining road can be saved, improve processing speed.
As previously discussed, 2 information processing unit 10A or system 1A according to embodiment, is looked into the distance in intersection etc.
The place of condition difference can support the safe driving of vehicle by using curved mirror.
In addition, in above-mentioned, the example being in during temporarily ceasing to vehicle is illustrated, but embodiment 2 also may be used
To be applied to the mobile vehicle of low speed (below predetermined speed).
(variation 1)
In embodiment 2, the size of the position according to the object in curved mirror and the object whenever necessary is sentenced
The situation for determining danger level is illustrated, but not limited to this.Identification part 13 can also identify the object appeared before one's eyes in curved mirror
Attribute, calculation section 101 is it is also contemplated that the attribute judges danger level.In this variation, to also considering and in curved mirror
The related attribute of movement speed possessed by the object appeared before one's eyes judges that the situation of danger level illustrates.
Figure 43 is to represent to represent that the risk prediction that the output processing part 102A of the variation 1 of embodiment 2 is carried out is handled
The definition graph of an example.In addition, pair describing identical statement, detailed description will be omitted with the same contents of Figure 40.In addition, in Figure 43
In show the variation 1 at the intersection of T words etc. output processing part 102A carry out risk prediction processing an example.This
Outside, based on it is low, in or high danger level vehicle control information can be as shown in Figure 40, therefore, omitted in Figure 43
Diagram.
As shown in figure 43, in the case where object is personage, risk prediction portion 1031A is determined as with similarly endangering during Figure 40
Dangerous degree.On the other hand, in the case where object is the movement speed bicycle faster than personage, motorcycle or automobile, danger
Dangerous prediction section 1031A be determined as according to movement speed be than object personage Shi Gao danger level.
As previously discussed, in this variation, output processing part 102A is according to the category of the object recognized by identification part 13
Property generate drive supporting information, output generated drive supporting information.
(variation 2)
In variation 1, to also considering the attribute related with movement speed possessed by the object appeared before one's eyes in curved mirror
Situation to judge danger level is illustrated, but not limited to this.It is dangerous in the case that the object appeared before one's eyes in curved mirror is people
Prediction section 1031A can also also consider the attribute related with the age of the people to judge danger level.Below using the situation as change
Shape example 2 illustrates.
Figure 44 is an example of risk prediction processing that the output processing part 102A for the variation 2 for representing embodiment 2 is carried out
Definition graph.In addition, pair describing identical statement, detailed description will be omitted with the same contents of Figure 40.In addition, show in Figure 44
Go out an example of risk prediction processing that the output processing part 102A of the variation 2 at the intersections such as T words is carried out.In addition, it is based on
It is low, in or high danger level vehicle control information can be as shown in Figure 40, therefore, figure is also omited in Figure 44
Show.
As shown in figure 44, in the case where object is personage and the personage is child or old man, risk prediction portion 1031A
It is determined as and the same danger levels of Figure 40.On the other hand, it is personage in object and the personage is other than child or old man
Other people in the case of, danger level high when being child or old man that risk prediction portion 1031A is determined as than personage.
As previously discussed, in this variation, output processing part 102A is in the attribute of the object recognized by identification part 13
In the case of being personage, become drive supporting information according to the information related with the age of the object recognized by identification part 13
Change to generate the drive supporting information, export the drive supporting information generated.It is pair related with the age in addition, in above-mentioned
Information be illustrated, but the information related with the age can also be the age or year of personage for the example of personage's generation
Generation.
(variation 3)
In the case of being people to the object appeared before one's eyes in curved mirror in variation 2, it is also contemplated that related with the age of the people
Attribute judge that the situation of danger level is illustrated, but not limited to this.The object appeared before one's eyes in curved mirror is the situation of people
Under, risk prediction portion 1031A can also also consider whether the people is walkinging this on one side see the portable terminals such as smart phone while
One attribute judges danger level.Specifically, do not notice that action is non-to see front action.For example, non-see that front action includes note
Depending on the portable terminals such as smart phone or books etc..It is illustrated below using the situation as variation 3.
Figure 45 is an example of risk prediction processing that the output processing part 102A for the variation 3 for representing embodiment 2 is carried out
Definition graph.In addition, pair describing identical statement, detailed description will be omitted with the same contents of Figure 40.In addition, show in Figure 45
Go out an example of risk prediction processing that the output processing part 102A of the variation 1 at the intersection of T words etc. is carried out.In addition, base
In it is low, in or high danger level vehicle control information can be as shown in Figure 40, therefore, figure is also omited in Figure 45
Show.
As shown in figure 45, it is personage in object and the personage is not carrying out looking at portable terminal and carrying out mobile action
It is non-see that the front action i.e. personage is not walkinged on one side look at portable terminal while in the case of, risk prediction portion 1031A judges
Same danger level during for Figure 40.On the other hand, it is personage in object and the personage is carrying out non-seeing front action
In the case that i.e. the personage is walkinged on one side look at portable terminal while, risk prediction portion 1031A is determined as than the personage not
Carry out it is non-see front action when high danger level.In addition, in above-mentioned, to not noticing that action is non-to see front action
Example is illustrated, though do not notice that action can also be looked at the front of personage but just look at side on the upper side or partial below
Action or watch the action of the specific object such as the perambulator for being located at the front of personage or ball attentively.
As previously discussed, in this variation, output processing part 102A is in the attribute of the object recognized by identification part 13
In the case of being personage, whether do not paid attention to taking action according to the object recognized by identification part 13 and believe drive supporting
Breath changes to generate the drive supporting information, exports the drive supporting information generated.
(embodiment 3)
Curved mirror can also be present in looking into the distance other than the T words road illustrated in embodiment 1 and 2 or curve turning road
The place of prestige condition difference.
Such as the place of the near exit in vehicle accommodation road, the possibility height that vehicle does not slow down and pop out next,
The garage entrance of shopping plaza, people or the emergent possibility of vehicle are high.In addition, for example in the entrance of shopping plaza, people's
It comes in and goes out more, therefore, the emergent possibility of people is high.In addition, the turning being connect with forthright for example easily raised speed in automobile
The place at angle etc., automobile be possible to not slow down and cross track scurry out that is to say pop out come.Such as these there are curved surfaces
The place for looking into the distance condition difference of mirror, the danger level that accident occurs are high.It that is to say, according to setting for the position of this vehicle or curved mirror
Place is put, since the condition that more places or the fireballing place of automobile etc. occurs in people is different, danger level is also different.
Then, in embodiment 3, to can by consider curved mirror setting place feature come in temporarily ceasing
Or information processing unit for being supported of safe driving of the vehicle during advancing etc. illustrates.In addition, in the following, it will
The vehicle of system or information processing unit equipped with embodiment 3 is known as this vehicle.
[composition of system 1B]
Figure 46 is the block diagram of an example of the composition for the system 1B for representing embodiment 3.In addition, pair same with Fig. 1 and Figure 26
The element of sample marks identical label, detailed description will be omitted.
System 1B shown in Figure 46 is compared with the system 1 of embodiment 1, acquisition unit 11B, detection process portion 12B and letter
The composition for ceasing processing unit 10B is different.In addition, system 1B is also equipped on the vehicle such as automobile, Neng Goutong in the same manner as system 1
Cross the drive supporting information that this i.e. vehicle of the vehicle is exported using curved mirror.
In the present embodiment, in order to prop up the safe driving of this vehicle in temporarily ceasing or during advancing
It helps, information processing unit 10B is also contemplated for the feature of the setting place of curved mirror.
[acquisition unit 11B]
Acquisition unit 11B is obtained in the same manner as image acquiring section 11 and is represented to pass through the shooting for the filming apparatus for being equipped on vehicle
The image information of obtained image.Acquisition unit 11B can also obtain cartographic information.Cartographic information refers to represent geographic position
The information of the information of the state of affairs of the dynamic change at place or geographic static state.Cartographic information includes representing the traffic thing on map
Therefore the addition cartographic information of at least one of congestion, construction, pavement state and meteorology.
In addition, acquisition unit 11B compared with the image acquiring section 11 of embodiment 1, can obtain cartographic information this point not
Together, other aspects are identical.In acquisition unit 11B, obtain detecting part 111 and cartographic information is obtained by communication etc., remembered
Record is in image recording portion 112.
[detection process portion 12B]
Detection process portion 12B detects curved mirror and the object appeared before one's eyes in curved mirror.In the present embodiment, at detection
Reason portion 12B is based on cartographic information or represents that the image of the image obtained by the shooting for the filming apparatus for being equipped on vehicle is believed
Breath detects the curved mirror of vehicle-surroundings.
As shown in figure 46, detection process portion 12B has test section 12B and identification part 13B.
＜ test section 12B ＞
Test section 12B detects curved mirror.More specifically, songs of the test section 12B based on image information detection vehicle-surroundings
Face mirror.For example, test section 12B detects vehicle using the identification information of curved mirror or the identifier of curved mirror based on image information
The curved mirror on periphery.In addition, the method for the curved mirror based on image information detection vehicle-surroundings is schemed with being detected in embodiment 1
The method of curved mirror as in is same.
In addition, test section 12B can also detect the curved mirror of vehicle-surroundings based on cartographic information.For example, test section 12B bases
In the setting place of the curved mirror represented by the position of vehicle and cartographic information, the curved mirror of vehicle-surroundings is detected.
＜ identification parts 13B ＞
Identification part 13B is known based on representing to obtain the image information of image by the shooting for the filming apparatus for being equipped on vehicle
The object do not appeared before one's eyes in curved mirror.In addition, identification part 13B uses the identification information of curved mirror or the identifier of curved mirror,
The ambient enviroment of curved mirror is identified based on image information.In this way, 13B identifications in identification part are in the image obtained by acquisition unit 11B
In the ambient enviroment of object in the curved mirror appeared before one's eyes and curved mirror.
In addition, acquisition unit 11B and detection process portion 12B can also be formed in the same manner as embodiment 1 and 2 at information
Manage device 10B.
[information processing unit 10B]
The information processing unit 10B of embodiment 3 is exported by using curved mirror in temporarily ceasing or during advancing
The drive supporting information of this vehicle.In the present embodiment, for the peace to this vehicle in temporarily ceasing or during advancing
Complete drive is supported, and information processing unit 10B is also contemplated for the feature of the setting place of curved mirror.As described above, this is because
It can be considered that the danger level different according to the setting place of the position of this vehicle or curved mirror.
More specifically, as shown in figure 46, information processing unit 10B has output processing part 102B, dead angle determination unit 103
And determination unit 104.Information processing unit 10B shown in Figure 46 is additional compared with the information processing unit 10 of embodiment 1
The composition of determination unit 104, the composition of output processing part 102B are different.Hereinafter, to determination unit 104 and output processing part 102B
The details of composition etc. illustrates.
104 ＞ of ＜ determination units
Figure 47 is the figure of an example that the function for the determination unit 104 for representing embodiment 3 is formed.
As shown in figure 47, it is dangerous to have feature determination unit 1041, the 1st danger level determination unit the 1042 and the 2nd for determination unit 104
Spend determination unit 1043.
《Feature determination unit 1041》
Feature determination unit 1041 judges the feature of the setting place of curved mirror.This feature includes the setting place of curved mirror
The state of road or the object of setting place by state.
More specifically, the feature of setting place of the feature determination unit 1041 based on cartographic information judgement curved mirror.For example,
The cartographic information on periphery of the feature determination unit 1041 based on curved mirror judges this feature.Feature determination unit 1041 can also be according to chasing after
Add cartographic information judgement road state or object by state.
In addition, feature determination unit 1041 can also judge the feature of the setting place of curved mirror based on image information.For example,
Feature determination unit 1041 can also judge feature based on the ambient enviroment of curved mirror.Feature determination unit 1041 can also be according to curved surface
Mirror ambient enviroment judgement road state or object by state.
Further, feature determination unit 1041 can also judge this feature based on the object appeared before one's eyes in curved mirror.In the feelings
Under condition, feature determination unit 1041 can also judge to appear before one's eyes the more of object or the object appeared before one's eyes in curved mirror in curved mirror
Widow is used as this feature.
In addition, feature determination unit 1041 can also also have a setting place acquisition unit, the setting place acquisition unit obtain with
The setting place for the curved mirror appeared before one's eyes in the image as obtained from the front to this vehicle is continuously shot in temporal sequence
The related information of feature.For the setting place acquisition unit, feature determination unit 1041 is not limited to the setting place acquisition unit
Situation, or detection process portion 12B or acquisition unit 11B have the setting place acquisition unit.
For example, setting place acquisition unit can be from GPS (the Global Positioning for the position for representing this vehicle
System, global positioning system) information acquirement image as obtained from the front to this vehicle is continuously shot in temporal sequence
The setting place of middle appeared before one's eyes curved mirror.In this case, setting place acquisition unit is according to acquired setting place, insertion
There is this earthquake of information, congestion information, road construction information, the scene of the accident and the information of road surface of traffic blackspot etc.
State figure obtains the information related with the feature of the setting place of curved mirror.
The setting place of curved mirror is characterized in the state of above-mentioned road.Specifically, the state of road is easily to send out
Make trouble thus actually have occurred the crowding of the road that the place of accident, the number of units of vehicle or congestion whether there is etc., fallen leaves,
Falling rocks, accumulated snow or the pavement behavior freezed etc. or whether there is construction etc..For example, the place being prone to accidents is that vehicle is special
With the near exit of road, the garage entrance of shopping plaza or entrance or the turn being connected with the forthright that automobile easily raises speed
Angle.
Figure 48 is the definition graph for an example for representing Dynamic Graph.Local Dynamic Graph is nearby to pay near one comprising predetermined location
Point Dynamic Graph, be to be overlapped this vehicle location periphery in the map datum of the fine as static informations such as road/atural objects
The multidate information that at every moment changes and cartographic information that High Level (advanced).Such as shown in figure 48, in local dynamic
In figure, the weight in the undermost basic data of the static information as information of road surface, lane information and three-dimensional structure etc.
It is laminated with semi-static information, Quasi dynamic data and dynamic data.Here, semi-static information is for example including traffic rule information, road
Road construction information and a wide range of weather information etc., Quasi dynamic data are for example including accident information, congestion information and small range
Weather information etc..Dynamic data ITS such as including nearby vehicle information, pedestrian's information and signal message
(Intelligent Transport Systems：Intelligent transportation system) predictive information etc..
In addition, setting place acquisition unit can also from the setting place obtained by GPS information, by by being set to this vehicle
The first-class shooting of vehicle-mounted pick-up obtained from the ambient enviroment of curved mirror appeared before one's eyes in image, obtain the setting field with curved mirror
The related information of feature.Further, the feature of the setting place of curved mirror may also mean that the object of above-mentioned setting place
Body by state.Specifically, the object of setting place by state be the presence or absence of discrepancy of people or vehicle or
Person's degree.
Here, the information related with the feature of the setting place of curved mirror refer to use as described above cartographic information or
The information for indicating whether the place to be prone to accidents that shooting image obtains.
Figure 49 A~Figure 49 E are an examples of the ambient enviroment of the curved mirror of embodiment 3.Ambient enviroment is, for example, such as Figure 49 A
The entrance of the commercial facilitys such as the shopping center shown in~Figure 49 E.In addition, the entrance of commercial facility has such as Figure 49 A~figure
Such variation shown in 49E, therefore, setting place acquisition unit can have the figure by the way that the entrance of commercial facility is appeared before one's eyes
The entrance identifier of commercial facility as obtained from study (Deep Learning etc.) as input.Setting place as a result,
The image of the ambient enviroment of curved mirror that acquisition unit can in the picture appear before one's eyes from expression, obtains the garage entrance for representing shopping plaza
Or the information in place that entrance etc. is prone to accidents.
Figure 50 is to represent the figure of an example of ground Figure 86 that the setting place acquisition unit of embodiment 3 uses.
Setting place acquisition unit can also after the ground Figure 86 for obtaining the current location for representing in Figure 50 this vehicle, from
The ambient enviroment for the curved mirror appeared before one's eyes in the picture obtains the information related with the feature of the setting place of curved mirror.This is because
Be located in the current location of this vehicle the commercial facilitys 860 such as the shopping center on ground Figure 86 shown in Figure 50 nearby then this vehicle
Current location be the parking garage of commercial facility 860 that can not obtain GPS information in the case of, this vehicle can be obtained
Current location be located at this case of commercial facility 860.This is because：Setting place acquisition unit can be from the picture as a result,
The ambient enviroment for the curved mirror appeared before one's eyes, higher precision area obtain the appearances such as the garage entrance for representing commercial facility 860 or outbound mouth
The information in the place of accident easily occurs.
In addition, in the case of being the outdoor parking space in shopping center that can obtain GPS information in the current location of this vehicle,
Setting place acquisition unit can also obtain the information related with the feature of the setting place of curved mirror from geography information.
《1st danger level determination unit 1042》
1st danger level determination unit 1042 is based on the feature of the setting place of curved mirror with being obtained by setting place acquisition unit
Related information, judgement represent that the 1st danger level of the height of the possibility of accident occurs for this vehicle.1st danger level determination unit
1042 both can be based on the setting place for the curved mirror that (acquirement) is judged with the current location according to this vehicle and cartographic information
Feature related information judges the 1st danger level, can also based on according to the ambient enviroment of curved mirror appeared before one's eyes in the picture
The related information of the feature of the setting place of the curved mirror of (acquirement) is judged to judge the 1st danger level.
First, the side of 1st danger level is judged the 1st danger level determination unit 1042 using Figure 51~Figure 54 based on geography information
An example of method illustrates.Figure 51~Figure 54 is to represent local of the 1st danger level determination unit 1042 of embodiment 3 for judgement
The figure of an example of Dynamic Graph.In Figure 51~Figure 54, identical label is marked to same element.
The one of the local Dynamic Graph 86t of the commuting time section obtained according to the current location of this vehicle is shown in Figure 51
Example.In Figure 51, road 861, road 862, super expressway 863 and road 864 are to represent this vehicle it is possible that the road of traveling
Road.In addition, in local Dynamic Graph 86t, it is set as the traffic accident comprising dynamic change, congestion information and represents across road
861 information of the more this case of people to commute.
In this case, the 1st danger level determination unit 1042 be set as mark hacures road 861 in this vehicle accident occurs
1st danger level of road 861 is determined as greatly by possibility highest.In addition, the 1st danger level determination unit 1042 is set as and road 861
Possibility time connected, that accident occurs by this vehicle in the road 862 of commercial facility 865 and mark hacures is high, by its 1st danger
During dangerous degree is determined as.1st danger level determination unit 1042 is set as super expressway 863 and road other than road 861 and 862
In 864 this vehicle occur accident possibility it is low, its 1st danger level is determined as small.
The large-scale local Dynamic Graph 86t connected when stopping obtained according to the current location of this vehicle is shown in Figure 521One
Example.In local Dynamic Graph 86t1In, traffic accident, congestion information as dynamic change are set as comprising expression in super expressway
The information of this case that a large amount of congestion and accident occurs in 863.In this case, the 1st danger level determination unit 1042 is set as
The possibility highest that accident occurs for this vehicle in the super expressway 863 of hacures is marked, the 1st danger level of super expressway 863 is sentenced
It is set to big.In addition, the 1st danger level determination unit 1042 be set as be connected with super expressway 863 mark hacures road 862 and
The possibility time height of accident occurs for this vehicle in road 864, during their the 1st danger level is determined as.1st danger level determination unit
1042 are set as the possibility that accident occurs for this vehicle in the road 862 other than super expressway 863, road 862 and road 864
It is low, its 1st danger level is determined as small.
In addition, the local of the period in the afternoon of the two-day weekend obtained according to the current location of this vehicle is shown in Figure 53
Dynamic Graph 86t2An example.In local Dynamic Graph 86t2In, traffic accident, congestion information as dynamic change are set as comprising table
Show the figure (graph) of the commercial facility 865 in shopping center etc. crowded period and represent be connected with commercial facility 865
The information of this case that a large amount of congestion and accident occurs at road 862.
In this case, the 1st danger level determination unit 1042 be set as mark hacures road 862 in this vehicle accident occurs
1st danger level of road 862 is determined as greatly by possibility highest.In addition, the 1st danger level determination unit 1042 be set as with road
The possibility time height of accident occurs for vehicle in the road 861 of 862 connected mark hacures, during its 1st danger level is determined as.The
1 danger level determination unit 1042 is set as this vehicle in the super expressway 863 other than road 861 and road 862 and road 864
The possibility of generation accident is low, their the 1st danger level is determined as small.
In addition, the local Dynamic Graph 86t when activity obtained according to the current location of this vehicle is shown in Figure 543One
Example.In addition, in local Dynamic Graph 86t3In, traffic accident, congestion information as dynamic change, be set as comprising represent with hair
It has given birth to this case that occur a large amount of congestion and accident at the connected road 862 of concert 866 of the activity such as concert
Information.
In this case, the 1st danger level determination unit 1042 be set as mark hacures road 862 in this vehicle accident occurs
Possibility highest, the 1st danger level of road 862 is determined as greatly.In addition, the 1st danger level determination unit 1042 be set as with road
The possibility time height of accident occurs for vehicle in the road 861 of the connected mark hacures in road 862, during its 1st danger level is determined as.
1st danger level determination unit 1042 is set as this vehicle in the super expressway 863 other than road 861 and road 862 and road 864
The possibility that accident occurs is low, their the 1st danger level is determined as small.
In this way, the 1st danger level determination unit 1042 can be based on this vehicle current location and represent because of current time, season
Section or current generation activity etc. and the traffic accident of dynamic change, congestion information geography information judge the 1st danger level.
Then, the 1st is judged based on the ambient enviroment for the curved mirror appeared before one's eyes in the picture to the 1st danger level determination unit 1042
An example of the method for danger level illustrates.Here, setting place acquisition unit is set as according to the curved mirror appeared before one's eyes in the picture
Ambient enviroment obtains the letter related with the feature of the setting place of the curved mirror in the entrance in shopping center or parking lot etc.
Breath.
1st danger level determination unit 1042 is being estimated as this vehicle according to the information related with the feature of the setting place of curved mirror
Parking lot of the current location for the commercial facility in shopping center etc. in the case of, be set as the possibility that accident occurs for this vehicle
To be moderate, during the 1st danger level is determined as.This is because a large amount of automobile that is at a stop in parking lot, the sheet in parking lot
The movement of vehicle to look into the distance condition poor.On the other hand, the 1st danger level determination unit 1042 is in basis and the setting place of curved mirror
The current location of the related information of feature and this vehicle is the entrance of commercial facilitys or the situations of garage entrance such as shopping center
Under, the possibility height that accident occurs for this vehicle is set as, the 1st danger level is determined as greatly.This is because：In the discrepancy of commercial facility
Mouthful, people on the move is more, and the possibility that this vehicle is collided with people is high.In addition, being the garage entrance because at the mall, usually have
There is slope, either motorcycle is possible to be accelerated or popped out after for the time being stopping automobile, the collision of this vehicle and people
Possibility is high.
《2nd danger level determination unit 1043》
2nd danger level determination unit 1043 is based on the 1st danger level judged by the 1st danger level determination unit 1042 and in curved mirror
The presence or absence of object appeared before one's eyes judges the 2nd danger level as the danger level predicted when this vehicle is advanced.
When with the object appeared before one's eyes in curved mirror, improved when this vehicle is advanced with the dangerous of object collision, because
This, the 2nd danger level that the judgement of the 2nd danger level determination unit 1043 takes into account.More specifically, the 2nd danger level determination unit
1043 in the case where the 1st danger level judged by the 1st danger level determination unit 1042 is small, with the object appeared before one's eyes in curved mirror
When, during the 2nd danger level is determined as, if the object do not appeared before one's eyes in curved mirror, the 2nd danger level is determined as low.2nd danger
In the case that dangerous degree determination unit 1043 is in the 1st danger level judged by the 1st danger level determination unit 1042 is, if in curved surface
During then 2 danger levels are determined as, if the object do not appeared before one's eyes in curved mirror, the 2nd danger level is sentenced for the object appeared before one's eyes in mirror
It is set to low.In addition, the 2nd danger level determination unit 1043 is big feelings in the 1st danger level judged by the 1st danger level determination unit 1042
Under condition, if with the object appeared before one's eyes in curved mirror, the 2nd danger level is determined as height, if do not appear before one's eyes in curved mirror
Object, then during the 2nd danger level is determined as.
Such as when in the position of this vehicle being the garage entrance of shopping plaza, at garage entrance, automobile or motorcycle have can
It can be accelerated or be popped out after stopping for the time being, therefore, be sentenced as described above by the 1st danger level determination unit 1042
It is set to during the 1st danger level is.2nd danger level determination unit 1043 is set as being located at storage when with the vehicle appeared before one's eyes in curved mirror
The collision possibility of this vehicle and vehicle near mouthful is high and is determined as the 2nd danger level.On the other hand, the 2nd danger level judges
In the vehicle do not appeared before one's eyes in curved mirror, this vehicle is low with the collision possibility of vehicle, and therefore, the 2nd danger level is sentenced in portion 1043
It is set to low.
In addition, for example the position of this vehicle for shopping plaza entrance near in the case of, near entrance, people
Discrepancy it is more, the probability of happening of accident can also increase, and therefore, be judged as described above by the 1st danger level determination unit 1042
It is big for the 1st danger level.2nd danger level determination unit 1043 is set as attached positioned at entrance when with the people to appear before one's eyes in curved mirror
This near vehicle and the collision possibility of people are high, and the 2nd danger level is determined as height.On the other hand, the 2nd danger level determination unit 1043
In the people not appeared before one's eyes in curved mirror, this vehicle is low with the collision possibility of people, therefore, during the 2nd danger level is determined as.
In addition, the 2nd danger level determination unit 1043 can also be based only upon the 1st danger judged by the 1st danger level determination unit 1042
Dangerous degree judges the 2nd danger level.In this case, if by the 1st danger level that the 1st danger level determination unit 1042 judges be it is big,
2nd danger level is determined as height by the 2nd danger level determination unit 1043.Equally, if being judged by the 1st danger level determination unit 1042
1st danger level is perhaps small in being, during the 2nd danger level is determined as by the 2nd danger level determination unit 1043 or low.
In this way, determination unit 104 is as in image obtained from being continuously shot in temporal sequence as the front to this vehicle
The feature of the setting place for the curved mirror appeared before one's eyes judges the 2nd danger level.
In addition, for example in feelings of the position of this vehicle for the place near limited highway or the entrance of expressway
Under condition, near entrance, the vehicle come out from vehicle accommodation road etc. is possible to not slow down, and the probability of happening of accident can also increase
Add, therefore, the 1st danger level can also be determined as greatly by the 1st danger level determination unit 1042.Also, the 2nd danger level determination unit 1043
When with the vehicle appeared before one's eyes in curved mirror, as the collision possibility of this vehicle and people height, it is determined as height i.e. the 2nd danger level
It can.On the other hand, for the 2nd danger level determination unit 1043 in the vehicle do not appeared before one's eyes in curved mirror, the collision of this vehicle and people can
Energy property is low, accordingly it is also possible to during the 2nd danger level is determined as.
In addition, in above-mentioned, to the example using the presence or absence of the object appeared before one's eyes in curved mirror in the judgement of the 2nd danger level
Son is illustrated, but can also in the judgement of the 2nd danger level using the type for the object appeared before one's eyes in curved mirror, quantity,
Number or closeness etc..
＜ output processing part 102B ＞
Output processing part 102B generates drive supporting information based on the feature judged by determination unit 104, and output is generated
Drive supporting information.Here, drive supporting information can both include the control instruction information of the movement of vehicle, can also include
The prompt message prompted to the occupant of vehicle.Prompt message can also include predicting the feature of the setting place according to curved mirror
The information prompted of danger.
Figure 55 is the figure of an example that the function for the output processing part 102B for representing embodiment 3 is formed.
In the present embodiment, as shown in figure 55, output processing part 102B has information generation unit 1032B and information output
Portion 1033B.
《Information generation unit 1032B》
Information generation unit 1032B generates drive supporting information based on the feature judged by determination unit 104.Information generation unit
1032B can also generate with the situation of road that is judged or object by the corresponding drive supporting information of situation.Example
Such as, information generation unit 1032B the road judged situation or object by situation be that can interfere the safety of vehicle
In the case of the situation of traveling, the drive supporting information for making vehicle deceleration, stopping or detour is generated.In addition, for example believe
Breath generating unit 1032B has according to appearing before one's eyes in curved mirror for being judged by determination unit 104 this case that object or in the curved mirror
In the number of object appeared before one's eyes, generate the drive supporting information for making vehicle deceleration, stopping or detour.
In addition, information generation unit 1032B can also generate drive supporting information according to the danger predicted by feature.
In addition, similary with embodiment 2, output processing part 102B can also be further in judgement to above-mentioned dead angle image
In the case of, generate drive supporting information.For example, output processing part 102B can also judged by dead angle determination unit 103 it is dead
During the image of angle, as drive supporting information, the information for vehicle to be made to temporarily cease certain period is generated.
《Information output part 1033B》
The drive supporting information that information output part 1033B outputs are generated by information generation unit 1032B.
Hereinafter, it is illustrated using an example that attached drawing handles the output processing part 102B formed in this way the outputs carried out.
Figure 56 and Figure 57 is to represent the definition graph of an example of output processing that the output processing part 102B of embodiment 3 is carried out.Scheming
The vehicle control information of this vehicle of the output processing part 102B outputs of the embodiment 3 at curve turning road is shown in 56
An example shows the vehicle control of this vehicle of the output processing part 102B outputs of the embodiment 3 at T words intersection in Figure 57
An example of information processed.In addition, in Figure 56 and Figure 57, on the basis of the 2nd danger level and its vehicle control information, according further to
The presence or absence of the 1st danger level based on curved mirror setting place and the object appeared before one's eyes in curved mirror are shown.
Therefore, output processing part 102B is in the case where this vehicle is located at curve turning road, based on being sentenced by the 2nd danger level
Determine the 2nd danger level of the judgement of portion 1043, generate and export vehicle control information for controlling vehicle as shown in figure 56 i.e.
It can.On the other hand, output processing part 102B is in the case where this vehicle is located at T words intersection, based on by the judgement of the 2nd danger level
The 2nd danger level that portion 1043 judges, generates and exports the vehicle control information for controlling vehicle as shown in figure 57.
In addition, relationship between the 2nd danger level and vehicle control information and the danger level illustrated with Figure 16 and Figure 17
Relationship between vehicle control information is likewise, therefore, omitting detailed description in this.
[action of system 1B]
Then, the information processing method of system 1B formed as described above is illustrated.Figure 58 is to represent embodiment 3
System 1B information processing method summary flow chart.Figure 59 is the information processing side for the system 1B for representing embodiment 3
The flow chart of the details of method.In addition, pair label identical with the same element label such as Figure 19, detailed description will be omitted.
As shown in figure 58, first, system 1B obtains cartographic information or represents to fill by the shooting for being equipped on vehicle
The image information (S30) for the image that the shooting put obtains.Then, system 1B is based on the cartographic information or figure obtained in S30
As information, the curved mirror (S31) of vehicle-surroundings is detected.Then, system 1B is based on cartographic information or image information to curved mirror
The feature of setting place judged (S32).Then, system 1B is based on the feature judged in S32, generation drive supporting letter
It ceases (S33).Also, the drive supporting information (S34) that system 1B outputs generate in S33.
More specifically, as shown in figure 59, first, system 1B carries out the processing with Figure 58 S30 illustrated.It is more specific and
Speech, system 1B are carried out obtaining cartographic information or be represented through the shooting for the filming apparatus for being equipped on vehicle in the processing of S30
The acquirement processing (S301) of the image information of obtained image.
Then, system 1B carries out the processing with Figure 58 S31 illustrated.More specifically, system 1B is in the processing of S31
In, carry out based on the cartographic information obtained in S301 or image information detection vehicle-surroundings curved mirror detection process
(S302).In addition, system 1B carries out the knowledge of object appeared before one's eyes in curved mirror based on the image information identification obtained in S302
Manage (S303) in other places.
Then, system 1B carries out the processing with Figure 58 S32 illustrated.More specifically, system 1B is in the processing of S32
In, carry out based on cartographic information or image information judgement curved mirror setting place feature feature determination processing
(S304).Then, system 1B carries out the 1st following danger level determination processing (S305)：Based on the setting place with curved mirror
The related information of feature, judgement represent that the 1st danger level of the height of the possibility of accident occurs for this vehicle.Then, system 1B into
The 2nd following danger level determination processing (S306) of row：Based on the 1st danger level judged in S305 and the object appeared before one's eyes in curved mirror
The presence or absence of, judge the 2nd danger level as the danger level predicted when this vehicle is advanced.In addition, about S304~S306 into
The details of capable processing is as described above, therefore, to omit detailed description in this.
In addition, in the processing of S33, system 1B is carried out according to the 2nd danger level generation drive supporting letter in S306 judgements
The information generation processing (S307) of breath.In addition, the details about the processing carried out in S307 is as described above, because
This, omits explanation in this.
Also, system 1B carries out the processing with Figure 57 S34 illustrated.More specifically, in S34, system 1B is carried out
The information output exported in the S307 drive supporting information generated etc. handles (S308).
[effect of embodiment 3 etc.]
As previously discussed, 3 information processing unit 10B or system 1B according to embodiment, is not only intersection, i.e.,
Make to look into the distance the place of condition difference in entrance of parking lot or commercial facility etc., also can judge vehicle by using curved mirror
Advance when danger level (the 2nd danger level).Further, it is possible to it is generated based on the 2nd danger level judged and exports vehicle
Drive supporting information, the safe driving therefore, it is possible to the vehicle to being equipped with information processing unit 10B are supported.
Specifically, the setting place in feature, that is, curved mirror of the setting place of curved mirror is intersection or parking
Field, the entrance of commercial facility, vehicle accommodation road near exit etc. look into the distance the place being prone to accidents of condition difference
In the case of, it is also possible to that accident occurs in this vehicle.It that is to say, in the case where the vehicle positioned at the proximal site is made to advance,
This vehicle is in the presence of the danger that accident can occur.Further, it is to look into the distance the easy generation of condition difference in the setting place of curved mirror
In the case of the place of accident, when appear before one's eyes in curved mirror have object when, the danger that accident occurs for this vehicle can further become
It is high.
Then, in the present embodiment, according to can be from temporarily ceasing or during advancing the obtained image of vehicle in
The presence or absence of the feature of the setting place for the curved mirror appeared before one's eyes and the object appeared before one's eyes in curved mirror, judgement is as when vehicle carries out
2nd danger level of the danger level of prediction.
In this way, the information processing unit 10B of embodiment 3 can be by using curved mirror come when judging to advance as vehicle
Danger level the 2nd danger level.
In addition, as described above, can also into according to can be from temporarily ceasing or during advancing the obtained figure of vehicle
The feature of the setting place of curved mirror appeared before one's eyes as in, the 2nd of the danger level that judgement is predicted when being carried out as vehicle are dangerous
Degree.
In addition, in the case that the vehicle in such as automatic Pilot has information processing unit 10B, which can be as above
It is described it is such judge the 2nd danger level using curved mirror, therefore, it is possible to carry out vehicle control according to the 2nd danger level judged
System.
As previously discussed, 3 information processing unit 10B or system 1B according to embodiment, in the field for looking into the distance condition difference
Institute can support the safe driving of vehicle by using curved mirror.
(variation 1)
In embodiment 3, feature to the setting place according to curved mirror and the object appeared before one's eyes in curved mirror have
Nothing is illustrated come the situation of the 2nd danger level of danger level predicted when judging and advancing as vehicle, but not limited to this.Identification
Portion 13B can also identify the attribute for the object appeared before one's eyes in curved mirror, and determination unit 104 is it is also contemplated that the attribute judges danger
Degree.In this variation, sentence to also considering the attribute related with movement speed possessed by the object appeared before one's eyes in curved mirror
The situation for determining danger level illustrates.
Figure 60 is the 2nd danger level judgement that the 2nd danger level determination unit 1043 of the variation 1 for representing embodiment 3 carries out
The definition graph of an example of processing.In addition, pair describing identical statement with the same contents of Figure 56 and Figure 57, omit specifically
It is bright.In addition, based on it is low, in the vehicle control information of either this vehicle of high the 2nd danger level can be Figure 56 Figure 57 institutes
Show that therefore, the illustration is omitted in Figure 60 like that.
As shown in figure 60, in the case where object is personage, the 2nd danger level determination unit 1043 is determined as and Figure 56 and figure
57 same 2nd danger levels.On the other hand, it is movement speed bicycle, motorcycle or the vapour faster than personage in object
In the case of vehicle, the 2nd danger level determination unit 1043 according to movement speed be determined as than object be the 2nd danger level of personage Shi Gao i.e.
It can.
As previously discussed, in this variation, the 2nd danger level determination unit 1043 judgement basis is recognized by identification part 13B
The attribute of object be allowed to the 2nd danger level changed.As a result, output processing part 102B can be identified according to by identification part 13B
To the attribute of object generate drive supporting information, output generated drive supporting information.
(variation 2)
In variation 1, to also considering the attribute related with movement speed possessed by the object appeared before one's eyes in curved mirror
Situation to judge the 2nd danger level is illustrated, but not limited to this.In the case that the object appeared before one's eyes in curved mirror is people,
The attribute related with the age of the people can also be also considered to judge the 2nd danger level.It is carried out below using the situation as variation 2
Explanation.
Figure 61 is the 2nd danger level judgement that the 2nd danger level determination unit 1043 of the variation 2 for representing embodiment 3 carries out
The definition graph of an example of processing.In addition, pair describing identical statement with the same contents of Figure 56 and Figure 57, omit specifically
It is bright.In addition, based on it is low, in the vehicle control information of either this vehicle of high the 2nd danger level can be Figure 56 Figure 57 institutes
Show like that, therefore, also omit diagram in figure 61.
As shown in Figure 61, in the case where object is personage and the personage is child or old man, the 2nd danger level determination unit
1043 are determined as and same 2nd danger level of Figure 56 and Figure 57.On the other hand, it is personage in object and the personage is
In the case of other people other than child or old man, the 2nd danger level determination unit 1043 is determined as that than personage be child or old man
When high the 2nd danger level.
As previously discussed, in this variation, the 2nd danger level determination unit 1043 is in the object recognized by identification part 13B
Attribute be personage in the case of, judgement according to the age of the object recognized by identification part 13B come be allowed to the change the 2nd endanger
Dangerous degree.As a result, output processing part 102B can in the case where the attribute of the object recognized by identification part 13B is personage,
The drive supporting information is generated to make drive supporting information change according to the age of the object recognized by identification part 13B,
The generated drive supporting information of output.In addition, in above-mentioned, pair information related with the age be the generation of personage example into
Explanation is gone, but the information related with the age can also be the age or age of personage.
(variation 3)
In variation 2 to the object appeared before one's eyes in curved mirror be people in the case of, also consider it is related with the age of the people
Attribute judge that the situation of the 2nd danger level is illustrated, but not limited to this.The object appeared before one's eyes in curved mirror is the feelings of people
Under condition, further, the 2nd danger level determination unit 1043 can also also consider the people, and whether to look at smart phone etc. on one side portable
Terminal one side walking this attribute judges the 2nd danger level.Specifically, do not notice that action is non-to see front action.It is for example, non-
See that front action includes watching the portable terminals such as smart phone or books etc. attentively.Illustrate below using the situation as variation 3.
Figure 62 is the 2nd danger level judgement that the 2nd danger level determination unit 1043 of the variation 3 for representing embodiment 3 carries out
The definition graph of an example of processing.In addition, pair describing identical statement with the same contents of Figure 56 and Figure 57, omit specifically
It is bright.In addition, based on it is low, in the vehicle control information of either this vehicle of high the 2nd danger level can be Figure 56 Figure 57 institutes
Show that therefore, the illustration is omitted in Figure 62 like that.
As shown in Figure 62, it is that personage and the personage carry out mobile action not carrying out looking at portable terminal in object
It is non-see that the front action i.e. personage is not walkinged on one side look at portable terminal while in the case of, the 2nd danger level determination unit 1043
Same 2nd danger level when being determined as with Figure 56 and Figure 57.On the other hand, object be personage and the personage
Carry out it is non-see that the front action i.e. personage is walkinged on one side look at portable terminal while in the case of, the 2nd danger level determination unit
1043 be determined as than the personage not carry out it is non-see front action when high the 2nd danger level.In addition, in above-mentioned, to not
The action of paying attention to is that the non-example for seeing front action is illustrated, although not noticing that action can also be before looking at personage
It side but looks at side on the upper side and either the action of partial below or watches the specific object such as the perambulator being located in front of personage or ball attentively
Action.
As previously discussed, in this variation, the 2nd danger level determination unit 1043 is in the object recognized by identification part 13B
Attribute be personage in the case of, judgement according to the object that is recognized by identification part 13B whether do not paid attention to action come
It is allowed to the 2nd danger level changed.As a result, output processing part 102B can be attribute in the object recognized by identification part 13B
In the case of being personage, make drive supporting according to whether the object recognized by identification part 13B is not being paid attention to action
Information change and generate and export the drive supporting information.
It as previously discussed, can be in intersection according to the information processing unit of the disclosure and information processing method etc.
The safe driving of vehicle is supported by using the curved mirror for being set to the place in the place poor etc. the condition of looking into the distance.Separately
Outside, the country other than Japan comes there is also curved mirror not too universal situation, but from the viewing to try to forestall traffic accidents etc.
It sees, is possible to set certain equipment for trying to forestall traffic accidents in the place for looking into the distance condition difference in the future.If here, use this public affairs
Information processing unit that each embodiment opened is related to etc. does not make equipment (such as the radar that equipment be configured is high price then
Deng), it will be able to the safety of traffic is improved with cheap curved mirror.
In addition, in above-mentioned embodiment etc., the shape of curved mirror be set as round shape or it is rectangular-shaped be illustrated,
But not limited to this.The shape of set curved mirror according to certain reasons and deformed mirror surface for convex concave situation or
The minute surface of curved mirror is also contained in for the situation of cloudy surface in the scope of the present disclosure.In this case or, in curved mirror
Reliability is imported in the identification for the object appeared before one's eyes, in the case where reliability is below threshold value, is set as what is appeared before one's eyes in curved mirror
The unreliable information of object, without using the information processing method of the disclosure.
More than, information processing unit of the disclosure and information processing method etc. are illustrated based on embodiment,
But the present invention is not limited to the embodiments.Without departing from the purport of the present invention, this field is implemented to present embodiment
Technical staff it is conceivable that various modifications obtained from mode or the inscape in the different embodiment of combination and structure
The mode built can also be included in the range of one or more modes of the present invention.For example, situation as described below is also wrapped
Containing in the present invention.
(1) specifically, above-mentioned each device is that have microprocessor, ROM, RAM, hard disk unit, display unit, key
The computer system of disk, mouse etc..Computer program is stored in the RAM or hard disk unit.It is pressed by the microprocessor
It is acted according to the computer program, each device realizes its function.Here, computer program be in order to realize predetermined function and group
It closes the command code of multigroup instruction for representing to be directed to computer and forms.
(2) part or all of the inscape of the above-mentioned each device of composition can also be by 1 system LSI (Large
Scale Integration：Large scale integrated circuit) it forms.System LSI be multiple constituting portion are integrated on 1 chip and
The super multi-functional LSI produced, specifically, being comprising microprocessor, ROM, RAM etc. and the computer system of composition.Institute
It states and computer program is stored in RAM.It is worked by the microprocessor according to the computer program, system LSI is real
Its existing function.
(3) part or all of the inscape of the above-mentioned each device of composition may be to have that each device can be loaded and unloaded on
IC card or monomer module.The IC card or the module are the departments of computer science being made of microprocessor, ROM, RAM etc.
System.The IC card or the module can also include above-mentioned super multi-functional LSI.By microprocessor according to computer program
It works, the IC card or the module realize its function.The IC card or the module can also have anti-distort.
(4) disclosure can also be above-mentioned shown method.Alternatively, it is also possible to being come real by these methods by computer
Existing computer program can also be the digital signal formed by the computer program.
(5) in addition, the disclosure can also be by the computer program or the digital signal record in computer-readable
The recording medium taken such as floppy disk, hard disk, CD-ROM, MO, DVD, DVD-ROM, DVD-RAM, BD (Blu-ray (registered trademark)
Disc), semiconductor memory etc..Alternatively, it is also possible to be recorded on the digital signal in these recording mediums.
(6) in addition, the disclosure can also be by the computer program or the digital signal via electrical communication lines, wireless
Or it wire communication line, is transmitted by network, data broadcasting of representative etc. of internet.
(7) in addition, the disclosure can also be the computer system for having microprocessor and memory, the memory record
There is above computer program, the microprocessor works according to the computer program.
(8) alternatively, it is also possible to by the way that described program or the digital signal record are turned in the recording medium
It moves or shifts described program or the digital signal via described network etc., calculated from there through independent others
Machine system is implemented.
[industrial availability]
The disclosure can be used in the vehicle-mounted camera for being equipped on vehicle, CAN for carrying out automatic Pilot
The system of (Controller Area Network, controller local area network) etc. or for carrying out at the information of drive supporting
Manage device or system etc..
Claims (19)
1. a kind of information processing unit, has：
Test section, based on the image information of image that expression is obtained by the shooting for the filming apparatus for being equipped on vehicle, detection
The curved mirror appeared before one's eyes in described image and the object appeared before one's eyes in the curved mirror；
Calculation section calculates the position of the object in the detected curved mirror；
Generating unit based on the position of the object in the curved mirror calculated, generates drive supporting information；And
Output section exports the drive supporting information generated.
2. information processing unit according to claim 1,
The generating unit is according to the object and the region in the curved mirror that is determined by the road appeared before one's eyes in described image
Between position relationship, generate the drive supporting information.
3. information processing unit according to claim 2,
The calculation section has：
Object space calculation section calculates the position of the object in the curved mirror；
Road surface region calculation section calculates the road surface region in the curved mirror；And
Central shaft calculation section, the central shaft in the road surface region calculated,
The generating unit is according to the position of the object calculated by the object space calculation section with being calculated by the central shaft
Position relationship between the central shaft that portion calculates generates the drive supporting information.
4. information processing unit according to claim 3,
The generating unit in the case where the object is located at the position more top than the central shaft in described image, use by generation
In the drive supporting information that the vehicle is made to hide the object.
5. information processing unit according to claim 2,
The generating unit is closed based on the position between the position calculated by the calculation section and the center line of the curved mirror
System, generates the drive supporting information.
6. information processing unit according to claim 5,
The generating unit,
(i) in the case where the curved mirror is from the point of view of the vehicle towards a left side, the object is located in described image than described
During the position that center line is kept right, alternatively,
(ii) in the case where the curved mirror is from the point of view of the vehicle towards the right side, the object is located in described image than institute
When stating the position that center line keeps left,
It generates that the vehicle is made to hide the drive supporting information of the object.
7. information processing unit according to claim 1,
The drive supporting information according to the vehicle under the situation advanced towards the direction of the curved mirror, by the object
The danger of position prediction of the body in the curved mirror exports.
8. information processing unit according to any one of claims 1 to 7,
The calculation section is also equipped with dimension of object calculation section, and the dimension of object calculation section is calculated on the basis of the curved mirror
The size of the object,
The generating unit is based on position of the object calculated in the curved mirror and the ruler of the object calculated
It is very little to generate the drive supporting information.
9. information processing unit according to claim 8,
The generating unit according to continuous at least two image in temporal sequence respectively in the respective ruler of the object appeared before one's eyes
Position of the very little change information and the object in the curved mirror, generates the drive supporting information.
10. information processing unit according to claim 1,
The drive supporting information includes the control instruction information of the movement of the vehicle.
11. information processing unit according to claim 1,
The drive supporting information includes the prompt message prompted to the occupant of the vehicle.
12. information processing unit according to claim 1,
Described information processing unit has the identification part of object that identification is appeared before one's eyes in the curved mirror.
13. information processing unit according to claim 12,
The generating unit generates the drive supporting information according to the attribute of the object recognized by the identification part.
14. information processing unit according to claim 13,
The generating unit in the attribute of the object recognized by the identification part in the case of personage, according to by described
Age of the object that identification part recognizes related information, makes the drive supporting information change and generates the drive supporting
Information.
15. information processing unit according to claim 13,
The generating unit is in the case where the attribute of the object recognized by the identification part is personage, according to by the knowledge
Whether the object that other portion recognizes, which is not paid attention to, is taken action, and is made the drive supporting information change and is generated the driving
Support information.
16. the information processing unit according to any one of claim 12~15,
The generating unit is further by the identification part at least one of continuous multiple images image in temporal sequence
The object recognized is not known in than image of at least one image in time series rearward by the identification part
In the case of being clipped to, the drive supporting information is generated.
17. information processing unit according to claim 16,
Position of the generating unit based on the object appeared before one's eyes in one image and the institute in the image rearward
It states object and this case is not recognized by the identification part, generate the drive supporting information.
18. a kind of information processing method, is performed using processor：
Based on the image information of image that expression is obtained by the shooting for the filming apparatus for being equipped on vehicle, detect in described image
In the curved mirror appeared before one's eyes and the object appeared before one's eyes in the curved mirror；
Calculate the position of the object in the detected curved mirror；
Based on the position of the object in the curved mirror calculated, drive supporting information is generated；
Export the drive supporting information generated.
19. a kind of program is to carry out the program that the computer of information processing method can be read, described program causes：
Based on the image information of image that expression is obtained by the shooting for the filming apparatus for being equipped on vehicle, detect in described image
In the curved mirror appeared before one's eyes and the object appeared before one's eyes in the curved mirror；
Calculate the position of the object in the detected curved mirror；
Based on the position of the object in the curved mirror calculated, drive supporting information is generated；
Export the drive supporting information generated.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2016254506 | 2016-12-27 | ||
| JP2016-254506 | 2016-12-27 | ||
| JP2017-152822 | 2017-08-08 | ||
| JP2017152822A JP6901937B2 (en) | 2016-12-27 | 2017-08-08 | Information processing equipment, information processing methods and programs |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108242182A true CN108242182A (en) | 2018-07-03 |
| CN108242182B CN108242182B (en) | 2021-11-05 |
Family
ID=60888119
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711390135.3A Active CN108242182B (en) | 2016-12-27 | 2017-12-21 | Information processing apparatus, information processing method, and recording medium |
Country Status (3)
| Country | Link |
|---|---|
| US (1) | US10788840B2 (en) |
| EP (1) | EP3343536B1 (en) |
| CN (1) | CN108242182B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115472027A (en) * | 2021-06-10 | 2022-12-13 | 欧温星球控股股份有限公司 | Remote assistance system and remote assistance method |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10788840B2 (en) * | 2016-12-27 | 2020-09-29 | Panasonic Intellectual Property Corporation Of America | Information processing apparatus, information processing method, and recording medium |
| US10304341B1 (en) * | 2018-02-09 | 2019-05-28 | International Business Machines Corporation | Vehicle and bicycle communication to avoid vehicle door crash accidents |
| US11163998B2 (en) | 2018-10-22 | 2021-11-02 | Woven Planet North America, Inc. | Systems and methods for automated image labeling for images captured from vehicles |
| JP2020095565A (en) * | 2018-12-14 | 2020-06-18 | トヨタ自動車株式会社 | Information processing system, program, and method for processing information |
| US11292398B2 (en) * | 2020-04-20 | 2022-04-05 | Hyundai Mobis Co., Ltd. | Apparatus for displaying forward blind spot situation |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006163637A (en) * | 2004-12-03 | 2006-06-22 | Fujitsu Ten Ltd | Driving supporting device |
| JP2006199055A (en) * | 2005-01-18 | 2006-08-03 | Advics:Kk | Vehicle running support apparatus |
| FR2898986A1 (en) * | 2006-03-24 | 2007-09-28 | Inrets | Obstacle detecting method for vehicle, involves providing central processing unit with acquisition module to receive and memorize impact data and video data, and processing data using unit to detect obstacle in road from geometry |
| CN101081608A (en) * | 2006-05-31 | 2007-12-05 | 爱信艾达株式会社 | Driving assist method and driving assist apparatus |
| CN101256713A (en) * | 2007-03-02 | 2008-09-03 | 富士通株式会社 | Driving assist system and vehicle-mounted apparatus |
| JP2009116527A (en) * | 2007-11-05 | 2009-05-28 | Mazda Motor Corp | Obstacle detecting apparatus for vehicle |
| JP2009211624A (en) * | 2008-03-06 | 2009-09-17 | Aisin Aw Co Ltd | Driving support device, driving support method, and computer program |
| US20100076621A1 (en) * | 2007-04-02 | 2010-03-25 | Panasonic Corporation | Safety driving support apparatus |
| JP2010191520A (en) * | 2009-02-16 | 2010-09-02 | Toyota Motor Corp | Surroundings monitor for vehicle |
| JP2013060057A (en) * | 2011-09-12 | 2013-04-04 | Toyota Motor Corp | Driving assist device |
| US20140085476A1 (en) * | 2011-06-02 | 2014-03-27 | Toyota Jidosha Kabushiki Kaisha | Vision support apparatus for vehicle |
| CN103733239A (en) * | 2011-11-01 | 2014-04-16 | 爱信精机株式会社 | Obstacle alert device |
| CN104015659A (en) * | 2013-02-28 | 2014-09-03 | 爱信精机株式会社 | Parking assist apparatus and parking assist method |
| CN104115199A (en) * | 2012-02-10 | 2014-10-22 | 三菱电机株式会社 | Driving assistance device and driving assistance method |
| CN104508722A (en) * | 2012-07-27 | 2015-04-08 | 歌乐株式会社 | Vehicle-mounted surrounding environment recognition device |
| CN105144261A (en) * | 2013-04-12 | 2015-12-09 | 丰田自动车株式会社 | Travel environment evaluation system, travel environment evaluation method, drive assist device, and travel environment display device |
| CN105378814A (en) * | 2013-07-19 | 2016-03-02 | 本田技研工业株式会社 | Vehicle travel safety device, vehicle travel safety method, and vehicle travel safety program |
| CN105787431A (en) * | 2015-01-13 | 2016-07-20 | 现代摩比斯株式会社 | Apparatus for safety-driving of vehicle |
| CN106156696A (en) * | 2015-03-30 | 2016-11-23 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6477464B2 (en) * | 2000-03-09 | 2002-11-05 | Donnelly Corporation | Complete mirror-based global-positioning system (GPS) navigation solution |
| US6727807B2 (en) * | 2001-12-14 | 2004-04-27 | Koninklijke Philips Electronics N.V. | Driver's aid using image processing |
| JP2007193577A (en) | 2006-01-19 | 2007-08-02 | Tokai Rika Co Ltd | Drive recorder utilizing device |
| US8482486B2 (en) * | 2009-04-02 | 2013-07-09 | GM Global Technology Operations LLC | Rear view mirror on full-windshield head-up display |
| US10614726B2 (en) * | 2014-12-08 | 2020-04-07 | Life Long Driver, Llc | Behaviorally-based crash avoidance system |
| JP6338626B2 (en) * | 2016-08-25 | 2018-06-06 | 株式会社Ｓｕｂａｒｕ | Vehicle display device |
| US10788840B2 (en) * | 2016-12-27 | 2020-09-29 | Panasonic Intellectual Property Corporation Of America | Information processing apparatus, information processing method, and recording medium |
| US10209718B2 (en) * | 2017-03-14 | 2019-02-19 | Starsky Robotics, Inc. | Vehicle sensor system and method of use |
| CN110826721A (en) * | 2018-08-09 | 2020-02-21 | 松下电器（美国）知识产权公司 | Information processing method and information processing system |
-
2017
- 2017-12-06 US US15/832,823 patent/US10788840B2/en active Active
- 2017-12-12 EP EP17206655.7A patent/EP3343536B1/en active Active
- 2017-12-21 CN CN201711390135.3A patent/CN108242182B/en active Active
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006163637A (en) * | 2004-12-03 | 2006-06-22 | Fujitsu Ten Ltd | Driving supporting device |
| JP2006199055A (en) * | 2005-01-18 | 2006-08-03 | Advics:Kk | Vehicle running support apparatus |
| FR2898986A1 (en) * | 2006-03-24 | 2007-09-28 | Inrets | Obstacle detecting method for vehicle, involves providing central processing unit with acquisition module to receive and memorize impact data and video data, and processing data using unit to detect obstacle in road from geometry |
| CN101081608A (en) * | 2006-05-31 | 2007-12-05 | 爱信艾达株式会社 | Driving assist method and driving assist apparatus |
| CN101256713A (en) * | 2007-03-02 | 2008-09-03 | 富士通株式会社 | Driving assist system and vehicle-mounted apparatus |
| US20100076621A1 (en) * | 2007-04-02 | 2010-03-25 | Panasonic Corporation | Safety driving support apparatus |
| JP2009116527A (en) * | 2007-11-05 | 2009-05-28 | Mazda Motor Corp | Obstacle detecting apparatus for vehicle |
| JP2009211624A (en) * | 2008-03-06 | 2009-09-17 | Aisin Aw Co Ltd | Driving support device, driving support method, and computer program |
| JP2010191520A (en) * | 2009-02-16 | 2010-09-02 | Toyota Motor Corp | Surroundings monitor for vehicle |
| US20140085476A1 (en) * | 2011-06-02 | 2014-03-27 | Toyota Jidosha Kabushiki Kaisha | Vision support apparatus for vehicle |
| JP2013060057A (en) * | 2011-09-12 | 2013-04-04 | Toyota Motor Corp | Driving assist device |
| CN103733239A (en) * | 2011-11-01 | 2014-04-16 | 爱信精机株式会社 | Obstacle alert device |
| CN104115199A (en) * | 2012-02-10 | 2014-10-22 | 三菱电机株式会社 | Driving assistance device and driving assistance method |
| CN104508722A (en) * | 2012-07-27 | 2015-04-08 | 歌乐株式会社 | Vehicle-mounted surrounding environment recognition device |
| CN104015659A (en) * | 2013-02-28 | 2014-09-03 | 爱信精机株式会社 | Parking assist apparatus and parking assist method |
| CN105144261A (en) * | 2013-04-12 | 2015-12-09 | 丰田自动车株式会社 | Travel environment evaluation system, travel environment evaluation method, drive assist device, and travel environment display device |
| CN105378814A (en) * | 2013-07-19 | 2016-03-02 | 本田技研工业株式会社 | Vehicle travel safety device, vehicle travel safety method, and vehicle travel safety program |
| CN105787431A (en) * | 2015-01-13 | 2016-07-20 | 现代摩比斯株式会社 | Apparatus for safety-driving of vehicle |
| CN106156696A (en) * | 2015-03-30 | 2016-11-23 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115472027A (en) * | 2021-06-10 | 2022-12-13 | 欧温星球控股股份有限公司 | Remote assistance system and remote assistance method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108242182B (en) | 2021-11-05 |
| US20180181139A1 (en) | 2018-06-28 |
| EP3343536A1 (en) | 2018-07-04 |
| EP3343536B1 (en) | 2021-02-17 |
| US10788840B2 (en) | 2020-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108241851A (en) | Information processing unit, information processing method and program | |
| CN108248602A (en) | Information processing unit, information processing method and program | |
| CN108242182A (en) | Information processing unit, information processing method and program | |
| Possatti et al. | Traffic light recognition using deep learning and prior maps for autonomous cars | |
| US11192544B2 (en) | Vehicle management system and commercial facility management system | |
| CN110210280B (en) | Beyond-visual-range sensing method, beyond-visual-range sensing system, terminal and storage medium | |
| CN109641589B (en) | Route planning for autonomous vehicles | |
| US8620032B2 (en) | System and method for traffic signal detection | |
| JP4656408B2 (en) | Ambient monitoring device for vehicles | |
| CN107449440A (en) | The display methods and display device for prompt message of driving a vehicle | |
| WO2023179028A1 (en) | Image processing method and apparatus, device, and storage medium | |
| TW201927610A (en) | Safety confirmation evaluating device, on-vehicle device, safety confirmation evaluation system having the two, safety confirmation evaluation method, and safety confirmation evaluation program | |
| KR101809088B1 (en) | Apparatus and method for forward collision warning | |
| JP7051326B2 (en) | Information processing equipment, information processing methods and programs | |
| CN114008408A (en) | Generating surface maps to improve navigation | |
| JP6953228B2 (en) | Information processing equipment, information processing methods and programs | |
| CN112537301B (en) | Driving reference object selection method and device for intelligent driving traffic carrier | |
| JP6901937B2 (en) | Information processing equipment, information processing methods and programs | |
| CN110287803B (en) | Method and system for identifying track and road sign | |
| JP2020190506A (en) | Traveling supporting device and computer program | |
| Yasin | Development of Automatic Number Plate Recognition Software and Journey Time Measurement | |
| Amri | Development of automatic number plate recognition software and journey time measurement/Amri Mohd Yasin |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |